chenxiawu_at_cs_dot_cornell.edu
I recieved my PhD in computer science at Cornell University. During PhD, I have been a visiting student researcher at Stanford AI Lab. I was advised by Ashutosh Saxena and collaborated with Silvio Savarese at Stanford.
I have been working on machine learning techniques in computer vision and robotics, and building artificial intelligence technologies for different products such as self-driving cars and smart home automations. My research focuses on machine learning of human activities and environments for robot perception. I built machine learning algorithms and systems to learn complex semantic, spatial and temporal structures from multiple domain sources such as cameras, RGB-D sensors, LIDAR sensors, PIR sensors, etc.
Our Watch-n-Patch work models human activities to learn about the complex relations in the activities when just given a completely unlabeled set of RGB-D videos. Discovering these complex relations are useful to recognizing actions, anticipating future actions, and detecting forgotten actions. We also provide a new large-scale RGB-D activity video dataset recorded by the new Kinect v2. Our algorithm has enabled a robot, called watch-bot, to observe and remind people in case they forgot some tasks in the house, thus making their life easier.
Papers published in ICRA 2016, CVPR 2015.
Media Covers: The Verge, IEEE SPECTRUM, New Scientist, CNET.
We use human as a strong evidence rather than only object visuals in automatic extraction of the common semantic regions given a set of images. We propose an unsupervised fully connected CRF auto-encoder incorporating the rich object features and a novel human-object interaction representation. We propose an efficient learning and inference algorithm to allow the full connectivity of the CRF with the auto-encoder, that establishes pairwise relations on all pairs of the object proposals in the dataset. Moreover, the auto-encoder learns the parameters from the data itself rather than supervised learning or manually assigned parameters in the conventional CRF. In the extensive experiments on four datasets, we show that our approach is able to extract the common objects more accurately than the state-of-the-art co-segmentation algorithms.
We present an algorithm that produces hierarchical labelings of a scene, following is-part-of and is-type-of relationships. We encode hierarchical labeling constraints into the conditional random field while keeping inference tractable. Our model thus predicts different specificities in labeling based on its confidence, thus lets the robot hedge its bet. In extensive experiments, both offline on standard datasets as well as in online robotic experiments (deployed in PR2 robot), we show that our model achieves good labeling performance as well as the success rate for robotic tasks.
Paper published in RSS 2014, invited talk in AAAI 2015.
We design learning based efficient binary feature coding and effective sparse representation for real-time and large-scale visual recognition problems such as object tracking, video copy detection, content-based image retrieval. We propose convolutional treelets binary feature retrieval based keypoint recognition approach, which can fast estimate the keypoint label and the pose for a query patch simultaneously. We proposed to converting global video features into binary codes by random projection and present a fast video copy retrieval method. We consider representation ability, discriminative power and efficiency of data representations in sparse coding to improve the image classification accuracy.
Papers published in IJCAI 2013, ECCV 2012, TRECVID 2011, IEEE TKDE, IEEE SMC.
Human Centred Object Co-Segmetation
Chenxia Wu, Jiemi Zhang, Ashutosh Saxena, Silvio Savarese.
Cornell Tech Report, 2016.
[ARXIV]
Watch-Bot: Unsupervised Learning for Reminding Humans of Forgotten Actions
Chenxia Wu, Jiemi Zhang, Bart Selman, Silvio Savarese, Ashutosh Saxena.
IEEE International Conference on Robotics and Automation (ICRA), 2016.
[PDF][PROJECT]
Watch-n-Patch: Unsupervised Understanding of Actions and Relations
Chenxia Wu, Jiemi Zhang, Silvio Savarese, Ashutosh Saxena.
Computer Vision and Pattern Recognition (CVPR), 2015.
[PDF][PROJECT]
Watch-n-Patch: Unsupervised Learning of Actions and Relations
Chenxia Wu, Jiemi Zhang, Bart Selman, Silvio Savarese, Ashutosh Saxena.
journal under review.
[ARXIV][PROJECT]
Hierarchical Semantic Labeling for Task-Relevant RGB-D Perception
Chenxia Wu, Ian Lenz, Ashutosh Saxena.
Robotics: Science and Systems (RSS), 2014.
[PDF][PROJECT]
Invited talk at AAAI 2015
Bilevel Visual Words Coding for Image Classification
Jiemi Zhang, Chenxia Wu, Deng Cai, Jianke Zhu.
International Joint Conference on Artificial Intelligence (IJCAI), 2013.
[PDF]
A Convolutional Treelets Binary Feature Approach to Fast Keypoint Recognition
Chenxia Wu, Jianke Zhu, Jiemi Zhang, Chun Chen, Deng Cai.
European Conference on Computer Vision (ECCV), 2012.
[PDF][VIDEO1][VIDEO2][CODE]
Treelets Binary Feature Retrieval for Fast Keypoint Recognition
Jianke Zhu, Chenxia Wu, Chun Chen, Deng Cai.
IEEE Transactions on Cybernetics, 2014. [PDF]
Semi-supervised Nonlinear Hashing Using Bootstrap Sequential Projection Learning
Chenxia Wu, Jianke Zhu, Deng Cai, Chun Chen, Jiajun Bu.
IEEE Transactions on Knowledge and Data Engineering (TKDE), 2012.
[PDF]
[CODE]
A Content-based Video Copy Detection Method with Randomly Projected Binary Features
Chenxia Wu, Jianke Zhu, Jiemi Zhang.
CVPR Workshop on Large-Scale Video Search and Mining, 2012. (TRECVID 2011 CCD task summary)
[PDF]
Sparse Poisson Coding for High Dimensional Document Clustering
Chenxia Wu, Haiqin Yang, Jianke Zhu, Jiemi Zhang, Irwin King, Michael R Lyu.
IEEE International Conference on Big Data, 2013.
[PDF]
Unsupervised Face-Name Association via Commute Distance
Jiajun Bu, Xu Bin, Chenxia Wu, Chun Chen, Jianke Zhu, Deng Cai, Xiaofei He.
ACM International Conference on Multimedia (ACM-MM), 2012.
[PDF]
Search Web Images Using Objects, Backgrounds and Conditions
Jiemi Zhang, Chenxia Wu, Deng Cai.
ACM International Conference on Multimedia (ACM-MM), 2012.
[PDF]
Cornell university fellowship, 2013
Chu Kochen Scholarship, highest honor at Zhejiang university, 2012
Google excellence scholarship, 2012
Outstanding postgraduates of Zhejiang province, 2013
Outstanding postgraduates of Zhejiang university, 2013
National scholarship for graduate students, 2012
First-class award of honor of Zhejiang university, 2012