Zhou Ren (任洲) Senior Research Scientist (CV, Google Scholar, Linkedin) Snap Inc., Venice, CA Email: zhou.ren [at] snapchat.com Contact me if you are interested in internship@Snap.

About me

• I am a technical innovator, actively seeking solutions to build and enhance real-world products. At Snap, I am leading efforts of applying visual understanding to support our 1). creative content creation, 2). content monetization, and 3). content security. I am the Tech Lead of two major production innovations, which handle multi-modal signals such as visual, textual, and audio inputs; involve many research scientists and engineers; cooperate with multiple engineering teams; and are building applications on both frontend and backend for Snap.

• I am an active researcher. My research interests include Computer Vision, Multimedia, Natural Language Processing and Machine Learning. I have worked on multi-modal joint understanding, video understanding, image captioning, hand gesture understanding, object detection, reinforcement learning, and adversarial machine learning, etc. I have obtained my Ph.D. in Computer Science at UCLA in 2016, advised by Professor Alan Yuille [Dissertation].

• Selected honors: (1) the 2nd place winner in NIPS 2017 Adversarial Attack and Defense Competition (among 107 teams); (2). the leading author of a work been nominated to “CVPR 2017 Best Student Paper Award”; (3). the leading author of the paper winning “IEEE Trans. on Multimedia 2016 Best Paper Award”; (4). I have developed the first part-based hand gesture recognition system using Kinect sensor with Nanyang Technological University and Microsoft Research Redmond (Demo1, Demo2).

News

• [01/18] One paper accepted by ICLR 2018. And a book chapter published in Springer on the topic of Adversarial Attacks and Defenses!

• [11/17] We have won the 2nd place in NIPS’17 Adversarial Defense Challenge among 107 teams; and were invited to present in NIPS’17 Competition Track (for details, please check our work).

• [08/17] Our work on hand gesture recognition is ranked the most cited paper in IEEE TMM since 2011!

• [07/17] One paper accepted by BMVC 2017 as an Oral!

• [07/17] Our paper on Reinforcement Learning-based Image Captioning has been nominated to the “Best Student Paper Award” at CVPR 2017!

• [03/17] One paper accepted by CVPR 2017 as an Oral!

• [01/17] Our TWO papers on hand gesture recognition are ranked the most cited paper && the 2nd most cited paper in ACM MM 2011!

• [06/16] Our paper “Robust Part-based Hand Gesture Recognition Using Kinect Sensor” has been selected to receive the 2016 IEEE Trans. on Multimedia Prize Paper Award (TMM Best Paper Award)!

Research

• My research interests lie in the fields of Computer Vision, Multimedia, Natural Language Processing, and Machine Learning. I have worked on multi-modal joint understanding, image captioning, hand gesture understanding, object detection, shape understanding, reinforcement learning, and adversarial machine learning, etc.

• My current research focuses include: 1) multi-modal joint understanding, 2) video understanding, 3) hand pose estimation, and 4) adversarial machine learning.

Publications

1. on Multi-modal Joint Representation Learning

2. on Adversarial Machine Learning

3. on Hand Gesture Recognition

4. on Object Modeling and Representation Learning

5. on Shape Representation and Shape Coding

6. on Medical Image Processing

Selected Patents

• System and method for robust hand gesture recognition using commodity depth sensor, Singapore provisional patent application, filed in 10/2011
  Co-invented with Junsong Yuan, Jingjing Meng

• Embedding space for images with multiple text labels, US patent application, filed in 01/2016
  Co-invented with Hailin Jin, Zhe Lin, and Chen Fang

• Modeling semantic concepts in an embedding space as distributions, US patent application, filed in 01/2016
  Co-invented with Hailin Jin, Zhe Lin, and Chen Fang

• Embedding-driven image captioning using deep reinforcement learning and lookahead beam search, US patent application, filed in 11/2016
  Co-invented with Xiaoyu Wang, Ning Zhang, Xutao Lv, and Jia Li

• Generating Data in a Messaging System for a Machine Learning Model, US patent application, filed in 12/2017
  Co-invented with Zehao Xue

• Query Matching to Media Collections in a Messaging System, US patent application, filed in 01/2018
  Co-invented with Roger Luo, Sushobhan Nayak, Xinran He, and Christophe Van Gysel

Teaching

• Teaching Associate of CS32 Introduction to Computer Science II (data structures and algorithms) with Prof. Carey Nachenberg
  [discussion materials]

• Teaching Assistant of CS31 Introduction to Computer Science I (C++ programming) with Prof. David Smallberg
  [discussion materials]

Professional Activities

• Program Committee of CVPR 2017, FG 2018, IJCAI 2018, ECAI 2018, ACM Multimedia 2018.

• Reviewer of FG 2016 2017 2018, WACV 2017, CVPR 2016 2017 2018, ICCV 2017, ECCV 2018, IJCAI 2018, etc.

• Reviewer of IEEE Trans on Image Processing; IEEE Trans on Circuits and Systems for Video Technology; IEEE Trans on Multimedia; IEEE Trans on Human-Machine Systems; Computer Vision and Image Understanding; Machine Vision and Application; Journal of Computer Science and Technology; etc.