Posts from 2017

Supercharge your Computer Vision models with the TensorFlow Object Detection API

Thursday, June 15, 2017

Crossposted on the Google Research Blog

At Google, we develop flexible state-of-the-art machine learning (ML) systems for computer vision that not only can be used to improve our products and services, but also spur progress in the research community. Creating accurate ML models capable of localizing and identifying multiple objects in a single image remains a core challenge in the field, and we invest a significant amount of time training and experimenting with these systems.
Detected objects in a sample image (from the COCO dataset) made by one of our models.
Image credit: Michael Miley, original image
Last October, our in-house object detection system achieved new state-of-the-art results, and placed first in the COCO detection challenge. Since then, this system has generated results for a number of research publications1,2,3,4,5,6,7 and has been put to work in Google products such as NestCam, the similar items and style ideas feature in Image Search and street number and name detection in Street View.

Today we are happy to make this system available to the broader research community via the TensorFlow Object Detection API. This codebase is an open source framework built on top of TensorFlow that makes it easy to construct, train and deploy object detection models.  Our goals in designing this system was to support state-of-the-art models while allowing for rapid exploration and research.  Our first release contains the following:
The SSD models that use MobileNet are lightweight, so that they can be comfortably run in real time on mobile devices. Our winning COCO submission in 2016 used an ensemble of the Faster RCNN models, which are are more computationally intensive but significantly more accurate.  For more details on the performance of these models, see our CVPR 2017 paper.

Are you ready to get started?
We’ve certainly found this code to be useful for our computer vision needs, and we hope that you will as well.  Contributions to the codebase are welcome and please stay tuned for our own further updates to the framework. To get started, download the code here and try detecting objects in some of your own images using the Jupyter notebook, or training your own pet detector on Cloud ML engine!

By Jonathan Huang, Research Scientist and Vivek Rathod, Software Engineer

The release of the Tensorflow Object Detection API and the pre-trained model zoo has been the result of widespread collaboration among Google researchers with feedback and testing from product groups. In particular we want to highlight the contributions of the following individuals:

Core Contributors: Derek Chow, Chen Sun, Menglong Zhu, Matthew Tang, Anoop Korattikara, Alireza Fathi, Ian Fischer, Zbigniew Wojna, Yang Song, Sergio Guadarrama, Jasper Uijlings, Viacheslav Kovalevskyi, Kevin Murphy

Also special thanks to: Andrew Howard, Rahul Sukthankar, Vittorio Ferrari, Tom Duerig, Chuck Rosenberg, Hartwig Adam, Jing Jing Long, Victor Gomes, George Papandreou, Tyler Zhu

  1. Speed/accuracy trade-offs for modern convolutional object detectors, Huang et al., CVPR 2017 (paper describing this framework)
  2. Towards Accurate Multi-person Pose Estimation in the Wild, Papandreou et al., CVPR 2017
  3. YouTube-BoundingBoxes: A Large High-Precision Human-Annotated Data Set for Object Detection in Video, Real et al., CVPR 2017 (see also our blog post)
  4. Beyond Skip Connections: Top-Down Modulation for Object Detection, Shrivastava et al., arXiv preprint arXiv:1612.06851, 2016
  5. Spatially Adaptive Computation Time for Residual Networks, Figurnov et al., CVPR 2017
  6. AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions, Gu et al., arXiv preprint arXiv:1705.08421, 2017
  7. MobileNets: Efficient convolutional neural networks for mobile vision applications, Howard et al., arXiv preprint arXiv:1704.04861, 2017

MobileNets: Open Source Models for Efficient On-Device Vision

Wednesday, June 14, 2017

Crossposted on the Google Research Blog

Deep learning has fueled tremendous progress in the field of computer vision in recent years, with neural networks repeatedly pushing the frontier of visual recognition technology. While many of those technologies such as object, landmark, logo and text recognition are provided for internet-connected devices through the Cloud Vision API, we believe that the ever-increasing computational power of mobile devices can enable the delivery of these technologies into the hands of our users, anytime, anywhere, regardless of internet connection. However, visual recognition for on device and embedded applications poses many challenges — models must run quickly with high accuracy in a resource-constrained environment making use of limited computation, power and space.

Today we are pleased to announce the release of MobileNets, a family of mobile-first computer vision models for TensorFlow, designed to effectively maximize accuracy while being mindful of the restricted resources for an on-device or embedded application. MobileNets are small, low-latency, low-power models parameterized to meet the resource constraints of a variety of use cases. They can be built upon for classification, detection, embeddings and segmentation similar to how other popular large scale models, such as Inception, are used.
Example use cases include detection, fine-grain classification, attributes and geo-localization.
This release contains the model definition for MobileNets in TensorFlow using TF-Slim, as well as 16 pre-trained ImageNet classification checkpoints for use in mobile projects of all sizes. The models can be run efficiently on mobile devices with TensorFlow Mobile.
Model Checkpoint
Million MACs
Million Parameters
Top-1 Accuracy
Top-5 Accuracy
Choose the right MobileNet model to fit your latency and size budget. The size of the network in memory and on disk is proportional to the number of parameters. The latency and power usage of the network scales with the number of Multiply-Accumulates (MACs) which measures the number of fused Multiplication and Addition operations. Top-1 and Top-5 accuracies are measured on the ILSVRC dataset.
We are excited to share MobileNets with the open source community. Information for getting started can be found at the TensorFlow-Slim Image Classification Library. To learn how to run models on-device please go to TensorFlow Mobile. You can read more about the technical details of MobileNets in our paper, MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications.

By Andrew G. Howard, Senior Software Engineer and Menglong Zhu, Software Engineer

MobileNets were made possible with the hard work of many engineers and researchers throughout Google. Specifically we would like to thank:

Core Contributors: Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam

Special thanks to: Benoit Jacob, Skirmantas Kligys, George Papandreou, Liang-Chieh Chen, Derek Chow, Sergio Guadarrama, Jonathan Huang, Andre Hentz, Pete Warden

Google Summer of Code 2017 statistics part 2

Tuesday, June 6, 2017

Now that Google Summer of Code (GSoC) 2017 is under way with students in their first full week of the coding period we wanted to bring you some more statistics on the 2017 program. Lots and lots of numbers follow:


Students are working with 201 organizations (the most we’ve ever had!) of which 39 are participating in GSoC for the first time.

Student Registrations

Over 20,651 students from 144 countries registered for the program, which is an 8.8% increase over the previous high for the program.

Project Proposals

4,764 students from 108 countries submitted a total of 7,089 project proposals.

Gender breakdown

11.4% of accepted students are women. We are always interested in making our programs and open source more inclusive. Please contact us if you know of organizations we should work with to spread the word about GSoC to underrepresented groups.


The 1,318 students accepted into the GSoC 2017 program hailed from 575 universities, of which 142 have students participating for the first time in GSoC.

Top 10 schools by students accepted for GSoC 2017 

University Name Country Accepted Students
International Institute of Information Technology, Hyderabad India 39
Birla Institute of Technology and Science, Pilani (BITS Pilani) India 37
Indian Institute of Technology, Kharagpur India 31
University of Moratuwa Sri Lanka 24
Delhi Technological University India 23
Birla Institute of Technology and Science Pilani, Goa Campus India 18
Indian Institute of Technology, Roorkee India 18
Indian Institute of Technology, Bombay India 15
LNM Institute of Information Technology India 15
TU Munich/Technische Universität München Germany 14

Another post with stats on our GSoC mentors will be coming soon!

Stephanie Taylor, Google Open Source

Google Summer of Code 2017 statistics part 1

Thursday, May 25, 2017

Since 2005 Google Summer of Code (GSoC) has been bringing new developers into the open source community every year. GSoC 2017 is the largest to date with 1,318 students from 72 countries accepted into the program who are working with a record 201 open source organizations this summer.

Students are currently participating in the Community Bonding phase of the program where they become familiar with the open source communities they will be working with. They also spend time learning the codebase and the community’s best practices so they can start their 12 week coding projects on May 30th.

Each year we like to share program statistics as we see GSoC continue to expand all over the world. This year there are three students that are the first to be accepted into GSoC from their home countries: Qatar, Tajikistan and Zimbabwe. A complete list of accepted students and their countries is below:

Country Students Country Students Country Students
Argentina 3 Ghana 1 Qatar 1
Armenia 1 Greece 29 Romania 11
Australia 6 Hungary 6 Russian Federation 54
Austria 13 India 569 Saudi Arabia 1
Bangladesh 2 Indonesia 2 Serbia 3
Belarus 3 Ireland 5 Singapore 10
Belgium 6 Israel 2 Slovak Republic 6
Bosnia and Herzegovina 1 Italy 23 Slovenia 2
Brazil 21 Jamaica 1 South Africa 2
Bulgaria 4 Japan 13 South Korea 8
Cameroon 8 Kazakhstan 1 Spain 19
Canada 27 Kenya 1 Sri Lanka 54
China 49 Latvia 1 Sweden 8
Colombia 1 Lithuania 2 Switzerland 5
Costa Rica 1 Macedonia 1 Taiwan 1
Croatia 1 Mexico 1 Tajikistan 1
Czech Republic 6 Moldova 1 Turkey 11
Denmark 2 Netherlands 14 Ukraine 12
Ecuador 2 New Zealand 1 United Arab Emirates 1
Egypt 10 Nigeria 1 United Kingdom 16
Estonia 1 Pakistan 8 United States 126
Finland 4 Peru 1 Uruguay 1
France 20 Poland 19 Vietnam 4
Germany 55 Portugal 10 Zimbabwe 1

In our next GSoC statistics post we will delve deeper into the schools, gender breakdown, mentors and registration numbers for the 2017 program.

Stephanie Taylor, Google Open Source

Open sourcing the Firebase SDKs

Wednesday, May 17, 2017

Today, at Google I/O 2017, we are pleased to announce that we are taking our first steps towards open sourcing our client libraries. By making our SDKs open, we’re aiming to show our commitment to greater transparency and to building a stronger developer community. To help further that goal, we’ll be using GitHub as a core part of our own toolchain to enable all of you to contribute as well. As you find issues in our code, from inconsistent style to bugs, you can file issues through the standard GitHub issue tracker. You can also find our project in the Google Open Source directory. We’re really looking forward to your pull requests!

What’s open?

We’re starting by open sourcing several products in our iOS, JavaScript, Java, Node.js and Python SDKs. We'll be looking at open sourcing our Android SDK as well. The SDKs are being licensed under Apache 2.0, the same flexible license as existing Firebase open source projects like FirebaseUI.

Let's take a look at each repo:

Firebase iOS SDK 4.0

With the launch of the Firebase iOS 4.0 SDKs we have made several improvements to the developer experience, such as more idiomatic API names for our Swift users. By open sourcing our iOS SDKs we hope to provide an additional avenue for you to give us feedback on such features. For this first release we are open sourcing our Realtime Database, Auth, Cloud Storage and Cloud Messaging (FCM) SDKs, but going forward we intend to release more.

Because we aren't yet able to open source some of the Firebase components, the full product build process isn't available. While you can use this repo to build a FirebaseDev pod, our libraries distributed through CocoaPods will continue to be static frameworks for the time being. We are continually looking for ways to improve the developer experience for developers, however you integrate.

Our GitHub README provides more details on how you build, test and contribute to our iOS SDKs.

Firebase JavaScript SDK 4.0

We are excited to announce that we are open sourcing our Realtime Database, Cloud Storage and Cloud Messaging (FCM) SDKs for JavaScript. We’ll have a couple of improvements hot on the heels of this initial release, including open sourcing Firebase Authentication. We are also in the process of releasing the source maps for our components, which we expect would really improve the debuggability of your app.

Our GitHub repo includes instructions on how you can build, test and contribute.

Firebase Admin SDKs


We are happy to announce that all three of our Admin SDKs for accessing Firebase on privileged environments are now fully open source, including our recently-launched Python SDK. While we continue to explore supporting more languages, we encourage you to use our source as inspiration to enable Firebase for your environment (and if you do, we'd love to hear about it!)

We're really excited to see what you do with the updated SDKs - as always reach out to us with feedback or questions in the Firebase-Talk Google Group, on Stack Overflow, via the Firebase Support team, and now on GitHub for SDK issues and pull requests! And to read about the other improvements to Firebase that launched at Google I/O, head over to the Firebase blog.

By Salman Qadri, Firebase Product Manager

Open Source at Google I/O 2017

Tuesday, May 16, 2017

One of the best parts of Google I/O every year is the chance to meet with the developers and community organizers from all over the world. It's a unique opportunity to have candid one-on-one conversations about the products and technologies we all love.

This year, I/O features a Community Lounge for attendees to relax, hangout, and play with neat experiments and games. It also features several mini-meetups during which you can chat with Googlers on a variety of topics.

Chris DiBona and Will Norris from the Google Open Source Programs Office will be around Thursday and Friday to talk about anything and everything open source, including our student outreach programs and the new Google Open Source website. If you're at Google I/O this year, make sure to drop by and say hello. Find dates, times, and other details in the Community Lounge schedule.

By Josh Simmons, Google Open Source

OSS-Fuzz: Five months later, and rewarding projects

Monday, May 8, 2017

Five months ago, we announced OSS-Fuzz, Google’s effort to help make open source software more secure and stable. Since then, our robot army has been working hard at fuzzing, processing 10 trillion test inputs a day. Thanks to the efforts of the open source community who have integrated a total of 47 projects, we’ve found over 1,000 bugs (264 of which are potential security vulnerabilities).

Breakdown of the types of bugs we're finding.

Notable results

OSS-Fuzz has found numerous security vulnerabilities in several critical open source projects: 10 in FreeType2, 17 in FFmpeg, 33 in LibreOffice, 8 in SQLite 3, 10 in GnuTLS, 25 in PCRE2, 9 in gRPC, and 7 in Wireshark, etc. We’ve also had at least one bug collision with another independent security researcher (CVE-2017-2801). (Some of the bugs are still view restricted so links may show smaller numbers.)

Once a project is integrated into OSS-Fuzz, the continuous and automated nature of OSS-Fuzz means that we often catch these issues just hours after the regression is introduced into the upstream repository, before any users are affected.

Fuzzing not only finds memory safety related bugs, it can also find correctness or logic bugs. One example is a carry propagating bug in OpenSSL (CVE-2017-3732).

Finally, OSS-Fuzz has reported over 300 timeout and out-of-memory failures (~75% of which got fixed). Not every project treats these as bugs, but fixing them enables OSS-Fuzz to find more interesting bugs.

Announcing rewards for open source projects

We believe that user and internet security as a whole can benefit greatly if more open source projects include fuzzing in their development process. To this end, we’d like to encourage more projects to participate and adopt the ideal integration guidelines that we’ve established.

Combined with fixing all the issues that are found, this is often a significant amount of work for developers who may be working on an open source project in their spare time. To support these projects, we are expanding our existing Patch Rewards program to include rewards for the integration of fuzz targets into OSS-Fuzz.

To qualify for these rewards, a project needs to have a large user base and/or be critical to global IT infrastructure. Eligible projects will receive $1,000 for initial integration, and up to $20,000 for ideal integration (the final amount is at our discretion). You have the option of donating these rewards to charity instead, and Google will double the amount.

To qualify for the ideal integration reward, projects must show that:
  • Fuzz targets are checked into their upstream repository and integrated in the build system with sanitizer support (up to $5,000).
  • Fuzz targets are efficient and provide good code coverage (>80%) (up to $5,000). 
  • Fuzz targets are part of the official upstream development and regression testing process, i.e. they are maintained, run against old known crashers and the periodically updated corpora (up to $5,000).
  • The last $5,000 is a “l33t” bonus that we may reward at our discretion for projects that we feel have gone the extra mile or done something really awesome.
We’ve already started to contact the first round of projects that are eligible for the initial reward. If you are the maintainer or point of contact for one of these projects, you may also reach out to us in order to apply for our ideal integration rewards.

The future

We’d like to thank the existing contributors who integrated their projects and fixed countless bugs. We hope to see more projects integrated into OSS-Fuzz, and greater adoption of fuzzing as standard practice when developing software.

By Oliver Chang, Abhishek Arya (Security Engineers, Chrome Security), Kostya Serebryany (Software Engineer, Dynamic Tools), and Josh Armour (Security Program Manager)

Students, Start Your Engineerings!

Thursday, May 4, 2017

It’s that time again! Our 201 mentoring organizations have selected 1,318 the students they look forward to working with during the 13th Google Summer of Code (GSoC). Congratulations to our 2017 students and a big thank you to everyone who applied!

The next step for participating students is the Community Bonding period which runs from May 4th through May 30th. During this time, students will get up to speed on the culture and toolset of their new community. They’ll also get acquainted with their mentor and learn more about the languages or tools they will need to complete their projects. Coding commences May 30th.

To the more than 4,200 students who were not chosen this year - don’t be discouraged! Many students apply at least once to GSoC before being accepted. You can improve your odds for next time by contributing to the open source project of your choice directly; organizations are always eager for new contributors! Look around GitHub and elsewhere on the internet for a project that interests you and get started.

Happy coding, everyone!

By Cat Allman, Google Open Source

Saddle up and meet us in Texas for OSCON 2017

Wednesday, April 26, 2017

The Google Open Source team is getting ready to hit the road and join the open source panoply that is Open Source Convention (OSCON). This year the event runs May 8-11 in Austin, Texas and is preceded on May 6-7 by the free-to-attend Community Leadership Summit (CLS).
Program chairs at OSCON 2016, left to right:
Kelsey Hightower, Scott Hanselman, Rachel Roumeliotis.
Photo used with permission from O'Reilly Media.

You’ll find our team and many other Googlers throughout the week on the program schedule and in the expo hall at booth #401. We’ve got a full rundown of our schedule below, but you can swing by the expo hall anytime to discuss Google Cloud Platform, our open source outreach programs, the projects we’ve open-sourced including Kubernetes, TensorFlow, gRPC, and even our recently released open source documentation.

Of course, you’ll also find our very own Kelsey Hightower everywhere since he is serving as one of three OSCON program chairs for the second year in a row.

Are you a student, educator, project maintainer, community leader, past or present participant in Google Summer of Code or Google Code-in? Join us for lunch at the Google Summer of Code table in the conference lunch area on Wednesday afternoon. We’ll discuss our outreach programs which help open source communities grow while providing students with real world software development experience. We’ll be updating this blog post and tweeting with details closer to the date.

Without further ado, here’s our schedule of events:

Monday, May 8th (Tutorials)

Tuesday, May 9th (Tutorials)

Wednesday, May 10th (Sessions)
12:30pm Google Summer of Code and Google Code-in lunch

Thursday, May 11th (Sessions)

We look forward to seeing you deep in the heart of Texas at OSCON 2017!

By Josh Simmons, Google Open Source

Introducing tf-seq2seq: An Open Source Sequence-to-Sequence Framework in TensorFlow

Tuesday, April 11, 2017

Crossposted on the Google Research Blog

Last year, we announced Google Neural Machine Translation (GNMT), a sequence-to-sequence (“seq2seq”) model which is now used in Google Translate production systems. While GNMT achieved huge improvements in translation quality, its impact was limited by the fact that the framework for training these models was unavailable to external researchers.

Today, we are excited to introduce tf-seq2seq, an open source seq2seq framework in TensorFlow that makes it easy to experiment with seq2seq models and achieve state-of-the-art results. To that end, we made the tf-seq2seq codebase clean and modular, maintaining full test coverage and documenting all of its functionality.

Our framework supports various configurations of the standard seq2seq model, such as depth of the encoder/decoder, attention mechanism, RNN cell type, or beam size. This versatility allowed us to discover optimal hyperparameters and outperform other frameworks, as described in our paper, “Massive Exploration of Neural Machine Translation Architectures.”

A seq2seq model translating from Mandarin to English. At each time step, the encoder takes in one Chinese character and its own previous state (black arrow), and produces an output vector (blue arrow). The decoder then generates an English translation word-by-word, at each time step taking in the last word, the previous state, and a weighted combination of all the outputs of the encoder (aka attention [3], depicted in blue) and then producing the next English word. Please note that in our implementation we use wordpieces [4] to handle rare words.
In addition to machine translation, tf-seq2seq can also be applied to any other sequence-to-sequence task (i.e. learning to produce an output sequence given an input sequence), including machine summarization, image captioning, speech recognition, and conversational modeling. We carefully designed our framework to maintain this level of generality and provide tutorials, preprocessed data, and other utilities for machine translation.

We hope that you will use tf-seq2seq to accelerate (or kick off) your own deep learning research. We also welcome your contributions to our GitHub repository, where we have a variety of open issues that we would love to have your help with!

We’d like to thank Eugene Brevdo, Melody Guan, Lukasz Kaiser, Quoc V. Le, Thang Luong, and Chris Olah for all their help. For a deeper dive into how seq2seq models work, please see the resources below.

[1] Massive Exploration of Neural Machine Translation Architectures, Denny Britz, Anna Goldie, Minh-Thang Luong, Quoc Le
[2] Sequence to Sequence Learning with Neural Networks, Ilya Sutskever, Oriol Vinyals, Quoc V. Le. NIPS, 2014
[3] Neural Machine Translation by Jointly Learning to Align and Translate, Dzmitry Bahdanau, Kyunghyun Cho, Yoshua Bengio. ICLR, 2015
[4] Google’s Neural Machine Translation System: Bridging the Gap between Human and Machine Translation, Yonghui Wu, Mike Schuster, Zhifeng Chen, Quoc V. Le, Mohammad Norouzi, Wolfgang Macherey, Maxim Krikun, Yuan Cao, Qin Gao, Klaus Macherey, Jeff Klingner, Apurva Shah, Melvin Johnson, Xiaobing Liu, Łukasz Kaiser, Stephan Gouws, Yoshikiyo Kato, Taku Kudo, Hideto Kazawa, Keith Stevens, George Kurian, Nishant Patil, Wei Wang, Cliff Young, Jason Smith, Jason Riesa, Alex Rudnick, Oriol Vinyals, Greg Corrado, Macduff Hughes, Jeffrey Dean. Technical Report, 2016
[5] Attention and Augmented Recurrent Neural Networks, Chris Olah, Shan Carter. Distill, 2016
[6] Neural Machine Translation and Sequence-to-sequence Models: A Tutorial, Graham Neubig
[7] Sequence-to-Sequence Models,

By Anna Goldie and Denny Britz, Research Software Engineer and Google Brain Resident, Google Brain Team

Join the first POSSE Workshop in Europe

Monday, April 10, 2017

We are excited to announce that the Professors’ Open Source Software Experience (POSSE) is expanding to Europe! POSSE is an event that brings together educators interested in providing students with experience in real-world projects through participation in humanitarian free and open source software (HFOSS) projects.

Over 100 faculty members have attended past workshops and there is a growing community of instructors teaching students through contributions to HFOSS. This three-stage faculty workshop will prepare you to support student participation in open source projects. During the workshop, you will:

  • Learn how to support student learning within real-world project environments
  • Motivate students and cultivate their appreciation of computing for social good
  • Collaborate with instructors who have similar interests and goals
  • Join a community of educators passionate about HFOSS

Workshop Format

Stage 1: Starts May 8, 2017 with online activities. Activities will take 2-3 hours per week and include interaction with workshop instructors and participants.
Stage 2: The face-to-face workshop will be held in Bologna, Italy, July 1-2, 2017 and is a pre-event for the ACM ITiCSE conference. Workshop participants include the workshop organizers, POSSE alumni, and members of the open source community.
Stage 3: Online activities and interactions in small groups immediately following the face-to-face workshop. Participants will have support while involving students in an HFOSS project in the classroom.

How to Apply

If you’re a full-time instructor at an academic institution outside of the United States, you can join the workshop being held in Bologna, Italy, July 1-2, 2017. Please complete and submit the application by May 1, 2017. Prior work with FOSS projects is not required. English is the official language of the workshop. The POSSE workshop committee will send an email notifying you of the status of your application by May 5, 2017.

Participant Support

The POSSE workshop in Europe is supported by Google. Attendees will be provided with funding for two nights lodging ($225 USD per night) and meals during the workshop. Travel costs will also be covered up to $450 USD. Participants are responsible for any charges above these limits. At this time, we can only support instructors at institutions of higher education outside of the U.S. For faculty at U.S. institutions, the next POSSE will be in fall 2017 on the east coast of the U.S.

We look forward to seeing you at the POSSE workshop in Italy!

By Helen Hu, Open Source Programs Office

Noto Serif CJK is here!

Thursday, April 6, 2017

Crossposted from the Google Developers Blog

Today, in collaboration with Adobe, we are responding to the call for Serif! We are pleased to announce Noto Serif CJK, the long-awaited companion to Noto Sans CJK released in 2014. Like Noto Sans CJK, Noto Serif CJK supports Simplified Chinese, Traditional Chinese, Japanese, and Korean, all in one font.

A serif-style CJK font goes by many names: Song (宋体) in Mainland China, Ming (明體) in Hong Kong, Macao and Taiwan, Minchō (明朝) in Japan, and Myeongjo (명조) or Batang (바탕) in Korea. The names and writing styles originated during the Song and Ming dynasties in China, when China's wood-block printing technique became popular. Characters were carved along the grain of the wood block. Horizontal strokes were easy to carve and vertical strokes were difficult; this resulted in thinner horizontal strokes and wider vertical ones. In addition, subtle triangular ornaments were added to the end of horizontal strokes to simulate Chinese Kai (楷体) calligraphy. This style continues today and has become a popular typeface style.

Serif fonts, which are considered more traditional with calligraphic aesthetics, are often used for long paragraphs of text such as body text of web pages or ebooks. Sans-serif fonts are often used for user interfaces of websites/apps and headings because of their simplicity and modern feeling.

Design of '永' ('eternity') in Noto Serif and Sans CJK. This ideograph is famous for having the most important elements of calligraphic strokes. It is often used to evaluate calligraphy or typeface design.

The Noto Serif CJK package offers the same features as Noto Sans CJK:

  • It has comprehensive character coverage for the four languages. This includes the full coverage of CJK Ideographs with variation support for four regions, Kangxi radicals, Japanese Kana, Korean Hangul and other CJK symbols and letters in the Unicode Basic Multilingual Plane of Unicode. It also provides a limited coverage of CJK Ideographs in Plane 2 of Unicode, as necessary to support standards from China and Japan.

Simplified Chinese
Supports GB 18030 and China’s latest standard Table of General Chinese Characters (通用规范汉字表) published in 2013.
Traditional Chinese
Supports BIG5, and Traditional Chinese glyphs are compliant to glyph standard of Taiwan Ministry of Education (教育部國字標準字體).
Supports all of the kanji in  JIS X 0208, JIS X 0213, and JIS X 0212 to include all kanji in Adobe-Japan1-6.
The best font for typesetting classic Korean documents in Hangul and Hanja such as Humninjeongeum manuscript, a UNESCO World Heritage.
Supports over 1.5 million archaic Hangul syllables and 11,172 modern syllables as well as all CJK ideographs in KS X 1001 and KS X 1002
Noto Serif CJK’s support of character and glyph set standards for the four languages
  • It respects diversity of regional writing conventions for the same character. The example below shows the four glyphs of '述' (describe) in four languages that have subtle differences.
From left to right are glyphs of '述' in S. Chinese, T. Chinese, Japanese and Korean. This character means "describe".
  • It is offered in seven weights: ExtraLight, Light, Regular, Medium, SemiBold, Bold, and Black. Noto Serif CJK supports 43,027 encoded characters and includes 65,535 glyphs (the maximum number of glyphs that can be included in a single font). The seven weights, when put together, have almost a half-million glyphs. The weights are compatible with Google's Material Design standard fonts, Roboto, Noto Sans and Noto Serif (Latin-Greek-Cyrillic fonts in the Noto family).
Seven weights of Noto Serif CJK
    • It supports vertical text layout and is compliant with the Unicode vertical text layout standard. The shape, orientation, and position of particular characters (e.g., brackets and kana letters) are changed when the writing direction of the text is vertical.

    The sheer size of this project also required regional expertise! Glyph design would not have been possible without leading East Asian type foundries Changzhou SinoType Technology, Iwata Corporation, and Sandoll Communications.

    Noto Serif CJK is open source under the SIL Open Font License, Version 1.1. We invite individual users to install and use these fonts in their favorite authoring apps; developers to bundle these fonts with your apps, and OEMs to embed them into their devices. The fonts are free for everyone to use!

    Noto Serif CJK font download:
    Noto Serif CJK on GitHub:
    Adobe's landing page for this release:
    Source Han Serif on GitHub:

    By Xiangye Xiao and Jungshik Shin, Internationalization Engineering team

    Google hosts the Apache HBase community at HBaseCon West 2017

    Friday, March 31, 2017

    We’re excited to announce that Google will host and organize HBaseCon West 2017, the official conference for the Apache HBase community on June 12. Registration for the event in Mountain View, CA is free and the call for papers (CFP) is open through April 24. Seats are limited and the CFP closes soon, so act fast.

    Apache HBase is the original open source implementation of the design concepts behind Bigtable, a critical piece of internal Google data infrastructure which was first described in a 2006 research paper and earned a SIGOPS Hall of Fame award last year. Since the founding of HBase, its community has made impressive advances supporting massive scale with enterprise users including Alibaba, Apple, Facebook, and Visa. The community is fostering a rich and still-growing ecosystem including Apache Phoenix, OpenTSDB, Apache Trafodion, Apache Kylin and many others.

    Now that Bigtable is available to Google Cloud users through Google Cloud Bigtable, developers have the benefit of platform choices for apps that rely on high-volume and low-latency reads and writes. Without the ability to build portable applications on open APIs,  however, even that freedom of choice can lead to a dead end  something Google addresses through its investment in open standards like Apache Beam, Kubernetes and TensorFlow.

    To that end, Google’s Bigtable team has been actively participating in the HBase community. We’ve helped co-author the HBase 1.0 API and have standardized on that API in Cloud Bigtable. This design choice means developers with HBase experience don’t need to learn a new API for building cloud-native applications, ensures Cloud Bigtable users have access to the large Apache Hadoop ecosystem and alleviates concerns about long-term lock-in.

    We hope you’ll join us and the HBase community at HBaseCon West 2017. We recommend registering early as there is no registration available on site. As usual, sessions are selected by the HBase community from a pool reflecting some of the world’s largest and most advanced production deployments.

    Register soon or submit a paper for HBaseCon  remember, the CFP closes on April 24! We look forward to seeing you at the conference.

    By Carter Page and Michael Stack, Apache HBase Project Management Committee members

    A New Home for Google Open Source

    Tuesday, March 28, 2017

    Google Open Source logo
    Free and open source software has been part of our technical and organizational foundation since Google’s early beginnings. From servers running the Linux kernel to an internal culture of being able to patch any other team's code, open source is part of everything we do. In return, we've released millions of lines of open source code, run programs like Google Summer of Code and Google Code-in, and sponsor open source projects and communities through organizations like Software Freedom Conservancy, the Apache Software Foundation, and many others.

    Today, we’re launching, a new website for Google Open Source that ties together all of our initiatives with information on how we use, release, and support open source.

    This new site showcases the breadth and depth of our love for open source. It will contain the expected things: our programs, organizations we support, and a comprehensive list of open source projects we've released. But it also contains something unexpected: a look under the hood at how we "do" open source.

    Helping you find interesting open source

    One of the tenets of our philosophy towards releasing open source code is that "more is better." We don't know which projects will find an audience, so we help teams release code whenever possible. As a result, we have released thousands of projects under open source licenses ranging from larger products like TensorFlow, Go, and Kubernetes to smaller projects such as Light My Piano, Neuroglancer and Some are fully supported while others are experimental or just for fun. With so many projects spread across 100 GitHub organizations and our self-hosted Git service, it can be difficult to see the scope and scale of our open source footprint.

    To provide a more complete picture, we are launching a directory of our open source projects which we will expand over time. For many of these projects we are also adding information about how they are used inside Google. In the future, we hope to add more information about project lifecycle and maturity.

    How we do open source

    Open source is about more than just code; it's also about community and process. Participating in open source projects and communities as a large corporation comes with its own unique set of challenges. In 2014, we helped form the TODO Group, which provides a forum to collaborate and share best practices among companies that are deeply committed to open source. Inspired by many discussions we've had over the years, today we are publishing our internal documentation for how we do open source at Google.

    These docs explain the process we follow for releasing new open source projects, submitting patches to others' projects, and how we manage the open source code that we bring into the company and use ourselves. But in addition to the how, it outlines why we do things the way we do, such as why we only use code under certain licenses or why we require contributor license agreements for all patches we receive.

    Our policies and procedures are informed by many years of experience and lessons we've learned along the way. We know that our particular approach to open source might not be right for everyone—there's more than one way to do open source—and so these docs should not be read as a "how-to" guide. Similar to how it can be valuable to read another engineer's source code to see how they solved a problem, we hope that others find value in seeing how we approach and think about open source at Google.

    To hear a little more about the backstory of the new Google Open Source site, we invite you to listen to the latest episode from our friends at The Changelog. We hope you enjoy exploring the new site!

    By Will Norris, Open Source Programs Office

    The latest round of Google Open Source Peer Bonus winners

    Monday, March 27, 2017

    Google relies on open source software throughout our systems, much of it written by non-Googlers. We’re always looking for ways to say “thank you!” so 5 years ago we started asking Googlers to nominate open source contributors outside of the company who have made significant contributions to codebases we use or think are important. We’ve recognized more than 500 developers from 30+ countries who have contributed their time and talent to over 400 open source projects since the program’s inception in 2011.

    Today we are pleased to announce the latest round of awardees, 52 individuals we’d like to recognize for their dedication to open source communities. The following is a list of everyone who gave us permission to thank them publicly:

    Name Project Name Project
    Philipp Hancke Adapter.js Fernando Perez Jupyter & IPython
    Geoff Greer Ag Michelle Noorali Kubernetes & Helm
    Dzmitry Shylovich Angular Prosper Otemuyiwa Laravel Hackathon Starter
    David Kalnischkies Apt Keith Busch Linux kernel
    Peter Mounce Bazel Thomas Caswell matplotlib
    Yuki Yugui Sonoda Bazel Tatsuhiro Tsujikawa nghttp2
    Eric Fiselier benchmark Anna Henningsen Node.js
    Rob Stradling Certificate Transparency Charles Harris NumPy
    Ke He Chromium Jeff Reback pandas
    Daniel Micay CopperheadOS Ludovic Rousseau PCSC-Lite, CCID
    Nico Huber coreboot Matti Picus PyPy
    Kyösti Mälkki coreboot Salvatore Sanfilippo Redis
    Jana Moudrá Dart Ralf Gommers SciPy
    John Wiegley Emacs Kevin O'Connor SeaBIOS
    Alex Saveau FirebaseUI-Android Sam Aaron Sonic Pi
    Toke Hoiland-Jorgensen Flent Michael Tyson The Amazing Audio Engine
    Hanno Böck Fuzzing Project Rob Landley Toybox
    Luca Milanesio Gerrit Bin Meng U-Boot
    Daniel Theophanes Go programming language Ben Noordhuis V8
    Josh Snyder Go programming language Fatih Arslan vim-go
    Brendan Tracey Go programming language Adam Treat WebKit
    Elias Naur Go on Mobile Chris Dumez WebKit
    Anthonios Partheniou Google Cloud Datalab Sean Larkin Webpack
    Marcus Meissner gPhoto2 Tobias Koppers Webpack
    Matt Butcher Helm Alexis La Goutte Wireshark dissector for QUIC

    Congratulations to all of the awardees, past and present! Thank you for your contributions.

    By Helen Hu, Open Source Programs Office

    Dispatches from the latest Mercurial sprints

    Friday, March 24, 2017

    On March 10th-12th, the Mercurial project held one of its twice-a-year sprints in the Google Mountain View office. Mercurial is a distributed version control system, used by Google, W3C, OpenJDK and Mozilla among others. We had 40 developers in attendance, some from companies with large Mercurial deployments and some individual contributors who volunteer in their spare time.

    One of the major themes we discussed was user-friendliness. Mercurial developers work hard to keep the command-line interface backwards compatible, but at the same time, we would like to make progress by smoothing out some rough edges. We discussed how we can provide a better user interface for users to opt-in to without breaking the backwards compatibility constraint. We also talked about how to make Mercurial’s Changeset Evolution feature easier to use.

    We considered moving Mercurial past SHA1 for revision identification, to enhance security and integrity of Mercurial repositories in light of recent SHA1 exploits. A rough consensus on a plan started to emerge, and design docs should start to circulate in the next month or so.

    We also talked about performance, such as new storage layers that would scale more effectively and work better with clones that only contain a partial repository history, a key requirement for Mercurial adoption in enterprise environments with large repositories, like Google.

    If you are interested in finding out more about Mercurial (or perhaps you’d like to contribute!) you can find our mailing list information here.

    By Martin von Zweigbergk and Augie Fackler, Software Engineers

    An Upgrade to SyntaxNet, New Models and a Parsing Competition

    Wednesday, March 22, 2017

    Crossposted from the Google Research Blog

    At Google, we continuously improve the language understanding capabilities used in applications ranging from generation of email responses to translation. Last summer, we open-sourced SyntaxNet, a neural-network framework for analyzing and understanding the grammatical structure of sentences. Included in our release was Parsey McParseface, a state-of-the-art model that we had trained for analyzing English, followed quickly by a collection of pre-trained models for 40 additional languages, which we dubbed Parsey's Cousins. While we were excited to share our research and to provide these resources to the broader community, building machine learning systems that work well for languages other than English remains an ongoing challenge. We are excited to announce a few new research resources, available now, that address this problem.

    SyntaxNet Upgrade
    We are releasing a major upgrade to SyntaxNet. This upgrade incorporates nearly a year’s worth of our research on multilingual language understanding, and is available to anyone interested in building systems for processing and understanding text. At the core of the upgrade is a new technology that enables learning of richly layered representations of input sentences. More specifically, the upgrade extends TensorFlow to allow joint modeling of multiple levels of linguistic structure, and to allow neural-network architectures to be created dynamically during processing of a sentence or document.

    Our upgrade makes it, for example, easy to build character-based models that learn to compose individual characters into words (e.g. ‘c-a-t’ spells ‘cat’). By doing so, the models can learn that words can be related to each other because they share common parts (e.g. ‘cats’ is the plural of ‘cat’ and shares the same stem; ‘wildcat’ is a type of ‘cat’). Parsey and Parsey’s Cousins, on the other hand, operated over sequences of words. As a result, they were forced to memorize words seen during training and relied mostly on the context to determine the grammatical function of previously unseen words.

    As an example, consider the following (meaningless but grammatically correct) sentence:
    This sentence was originally coined by Andrew Ingraham who explained: “You do not know what this means; nor do I. But if we assume that it is English, we know that the doshes are distimmed by the gostak. We know too that one distimmer of doshes is a gostak." Systematic patterns in morphology and syntax allow us to guess the grammatical function of words even when they are completely novel: we understand that ‘doshes’ is the plural of the noun ‘dosh’ (similar to the ‘cats’ example above) or that ‘distim’ is the third person singular of the verb distim. Based on this analysis we can then derive the overall structure of this sentence even though we have never seen the words before.

    To showcase the new capabilities provided by our upgrade to SyntaxNet, we are releasing a set of new pretrained models called ParseySaurus. These models use the character-based input representation mentioned above and are thus much better at predicting the meaning of new words based both on their spelling and how they are used in context. The ParseySaurus models are far more accurate than Parsey’s Cousins (reducing errors by as much as 25%), particularly for morphologically-rich languages like Russian, or agglutinative languages like Turkish and Hungarian. In those languages there can be dozens of forms for each word and many of these forms might never be observed during training - even in a very large corpus.

    Consider the following fictitious Russian sentence, where again the stems are meaningless, but the suffixes define an unambiguous interpretation of the sentence structure:
    Even though our Russian ParseySaurus model has never seen these words, it can correctly analyze the sentence by inspecting the character sequences which constitute each word. In doing so, the system can determine many properties of the words (notice how many more morphological features there are here than in the English example). To see the sentence as ParseySaurus does, here is a visualization of how the model analyzes this sentence:
    Each square represents one node in the neural network graph, and lines show the connections between them. The left-side “tail” of the graph shows the model consuming the input as one long string of characters. These are intermittently passed to the right side, where the rich web of connections shows the model composing words into phrases and producing a syntactic parse. Check out the full-size rendering here.

    A Competition
    You might be wondering whether character-based modeling are all we need or whether there are other techniques that might be important. SyntaxNet has lots more to offer, like beam search and different training objectives, but there are of course also many other possibilities. To find out what works well in practice we are helping co-organize, together with Charles University and other colleagues, a multilingual parsing competition at this year’s Conference on Computational Natural Language Learning (CoNLL) with the goal of building syntactic parsing systems that work well in real-world settings and for 45 different languages.

    The competition is made possible by the Universal Dependencies (UD) initiative, whose goal is to develop cross-linguistically consistent treebanks. Because machine learned models can only be as good as the data that they have access to, we have been contributing data to UD since 2013. For the competition, we partnered with UD and DFKI to build a new multilingual evaluation set consisting of 1000 sentences that have been translated into 20+ different languages and annotated by linguists with parse trees. This evaluation set is the first of its kind (in the past, each language had its own independent evaluation set) and will enable more consistent cross-lingual comparisons. Because the sentences have the same meaning and have been annotated according to the same guidelines, we will be able to get closer to answering the question of which languages might be harder to parse.

    We hope that the upgraded SyntaxNet framework and our the pre-trained ParseySaurus models will inspire researchers to participate in the competition. We have additionally created a tutorial showing how to load a Docker image and train models on the Google Cloud Platform, to facilitate participation by smaller teams with limited resources. So, if you have an idea for making your own models with the SyntaxNet framework, sign up to compete! We believe that the configurations that we are releasing are a good place to start, but we look forward to seeing how participants will be able to extend and improve these models or perhaps create better ones!

    Thanks to everyone involved who made this competition happen, including our collaborators at UD-Pipe, who provide another baseline implementation to make it easy to enter the competition. Happy parsing from the main developers, Chris Alberti, Daniel Andor, Ivan Bogatyy, Mark Omernick, Zora Tung and Ji Ma!

    By David Weiss and Slav Petrov, Research Scientists