Jie's Exploration Fairyland

Give the machine a sensible Eye

I am a second-year CS master student at the University of Rochester, working with Prof. Chenliang Xu and Prof. Jiebo Luo. Prior to that, I received my B.S. degree in Electronic Information and master in Communication and Information System from Beijing Normal University.
My general research interest lies in computer vision, especially in video analysis, action recognition. [CV]


Weakly-supervised video actor-action segmentation

Project leader @University of Rochester, Oct. 2017 ~ Now
  • video-level weakly-supervised actor-action segmentation -- involve the tasks of video object segmentation and action recognition -- on A2D (unconstrained in-the-wild YouTube videos)
  • in the complete absence of pixel-wise annotation, achieve 72% mIoU of the best fully-supervised model on A2D
  • also achieve state-of-the-art performance on weakly-supervised video object segmentation on the YouTube-Object dataset
  • robust to real-world adverse conditions: background clutter, motion blur, low illumination, occlusion, out of view, and small scale, etc. [Demo]

Spatial-temporal action localization in AVA dataset

Full-time Research Intern (Mentor: Lijuan Wang) @Microsoft Research, Seattle, May. 2018 ~ Aug. 2018
  • Transfer the long-term modeling in successful speech recognition architectures to long duration action sequence prediction on AVA dataset.
  • Model multi-person contextual dependency by means of spatial action heatmap.
  • Replicate the faster-rcnn+I3D based model that wins the AVA Challenge 2018 at the CVPR workshop.


Pose estimation

Wentao Liu, Jie Chen, Cheng Li, Chen Qian, Xiao Chu, and Xiaolin Hu, A Cascaded Inception of Inception Network with Attention Modulated Feature Fusion for Human Pose Estimation, In Thirty-Second AAAI Conference on Artificial Intelligence, 2018. [PDF]

ROI Coding

Libao Zhang, Jie Chen, and Bingchang Qiu, Region of Interest Coding Based on Saliency Detection and Directional Wavelet for Remote Sensing Images, IEEE Geoscience and Remote Sensing Letters, online. [PDF]

ROI Joint-saliency

Jie Chen and Libao Zhang, Joint Multi-Image Saliency Analysis for Region of Interest Detection in Optical Multispectral Remote Sensing Images, Remote Sensing, vol. 8, pp. 461, May 2016. [PDF]

Saliency with directional wavelet

Libao Zhang, Jie Chen, and Bingchang Qiu, Region of interest extraction in remote sensing images by saliency analysis with the normal directional lifting wavelet transform, Neurocomputing, vol. 179, pp. 186-201, February 2016. [PDF]

Things I am facinated with

Be more patient to polish your every skill.

  • Coding in python, c++
  • Stack network using Pytorch, Caffe, Keras
  • Playing the Ukelele
  • Cooking Sichuan cuisine
  • Playing badminton and zumba dancing
  • Learn watercolor once I get time!!

Contact Me

Email: jchen121 at cs dot rochester dot edu
Address: Wegmans Hall 2th floor, University of Rochester, Rochester, NY 14627