DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
Instructor: Ben Ochoa
Email: bochoa at ucsd.edu
Office hours: W 8:00 PM-9:00 PM, EBU3B 3208
TA: Zak Murez
Email: zmurez at eng.ucsd.edu
Office hours: W and Th 3:00 PM-4:00 PM, EBU3B 4127
Note: when emailing the instructor or TA with questions about the class, please put "CSE 252C" in the subject line.
Class section ID: 903801
Lecture: MW 6:30 PM-7:50 PM, CENTR 105
Class discussion: Piazza
This course is dedicated to recent research on computer vision, including learning-based methods. Students enrolled in this course are required to present a paper in class and complete a project inspired by the paper.
When presenting a paper in class, follow the presentation guidelines provided by Professor Charles Elkan. One week prior to your presentation date, send a draft of your slides to the instructor and TA for review. The instructor and TA will provide you with comments to incorporate into your slides prior to your presentation in class. Immediately after your presentation, send the slides (PDF, two slides per page) to the instructor for publishing on the class website. You will be receive talk feedback from the instructor and TA.
All projects will follow specific guidelines, including preparation of a project proposal, draft project report, and final project report. The project should be at the frontier of current research, but need not necessarily advance the state of the field. For example, replicating the results of an innovative paper would be a good project. Projects must be closely inspired by one or two specific high quality papers and should have an experimental aspect. Project reports will be evaluated using these grading criteria.
Prerequisites: CSE 252A or CSE 252B.
Academic Integrity Policy: Integrity of scholarship is essential for an academic community. The University expects that both faculty and students will honor this principle and in so doing protect the validity of University intellectual work. For students, this means that all academic work will be done by the individual to whom it is assigned, without unauthorized aid of any kind.
Collaboration Policy: It is expected that you complete your academic assignments on your own and in your own words and code. The assignments have been developed by the instructor to facilitate your learning and to provide a method for fairly evaluating your knowledge and abilities (not the knowledge and abilities of others). So, to facilitate learning, you are authorized to discuss assignments with others; however, to ensure fair evaluations, you are not authorized to use the answers developed by another, copy the work completed by others in the past or present, or write your academic assignments in collaboration with another person. If the work you submit is determined to be other than your own, you will be reported to the Academic Integrity Office for violating UCSD's Policy on Integrity of Scholarship.
Grading: Course grades will be based 50% on presentations in class, and participation in class and online; and 50% on the project report.
Papers:
| Date | Presenters | Paper | Slides |
|---|---|---|---|
| April 10 | Xinxin Chen, Yifeng Bu, Shengyao Guo, and Wenhao Sheng | Deep Visual-Semantic Alignments for Generating Image Descriptions (A. Karpathy and L. Fei-Fei) [pdf] | |
| April 12 | Kwonjoon Lee, Nimish Srivastava, Songting Xu, and Ali Mirzaei | Generative Adversarial Nets (I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozairz, A. Courville, and Y. Bengio) [pdf] | |
| April 17 | Group meetings | ||
| April 19 | Group meetings | ||
| April 24 | Dhanesh Pradhan | A Human Activity Recognition System Using Skeleton Data from RGBD Sensors (E. Cippitelli, S. Gasparrini, E. Gambi, and S. Spinsante) [pdf] | |
| April 26 | Mingdong Wang and Yixin Yang | Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (S. Ren, K. He, R. Girshick, and J. Sun) [pdf] | |
| May 1 | Zhipeng Yan, Moyuan Huang, and Hao Jiang | DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs (L.C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, A.L. Yuille) [pdf] | |
| May 3 | Akshaya Purohit, Lenord Melvix, Xiaoyu Zhou, and Bolun Zhang | ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras (R. Mur-Artal and J.D. Tardos) [pdf] | |
| May 8 | Yuting Wang and Yuwei Wang | Video Paragraph Captioning Using Hierarchical Recurrent Neural Networks (H. Yu, J. Wang, Z. Huang, Y. Yang, W. Xu [pdf] | |
| May 10 | Anthony Thomas and Joni De Guzman | A minimal parameterization of the trifocal tensor (K. Nordberg) [pdf] | |
| May 15 | Adithya Seshasayee and Andrew Durnford | Globally Optimal Algorithms for Stratified Autocalibration (M. Chandraker, S. Agarwal, D. Kriegman, and S. Belongie) [pdf] | |
| May 17 | Justin Gorgen, Haifeng Huang, Hao-en Sung, and Yen-Ting Chen | VINet: Visual-Inertial Odometry as a Sequence-to-Sequence Learning Problem (R. Clark, S. Wang, H. Wen, A. Markham, and N. Trigoni) [pdf (preprint)] | |
| May 22 | Ishan Gupta and Shashank Tyagi | Spatial Transformer Networks (M. Jaderberg, K. Simonyan, A. Zisserman, and K. Kavukcuoglu) [pdf] | |
| May 24 | Mike Liu, Jing Wang, and Yuhang Ming | Struck: Structured Output Tracking with Kernels (S. Hare, A. Saffari, and P.H.S. Torr) [pdf] | |
| May 29 | No meeting (Memorial Day Observance) | ||
| May 31 | Harshita Mangal, Zhengqin Li, and Ji Dai | Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks (J.Y. Zhu, T. Park, P. Isola, and A.A. Efros) [pdf] | |
| June 5 |
Project presentations Kwonjoon Lee, Nimish Srivastava, Songting Xu, and Ali Mirzaei: A Comparative Study of Different GANs Dhanesh Pradhan: Human Activity Recognition using Skeleton Data from RGBD Sensors Mingdong Wang and Yixin Yang: Object Detection for Self Driving using Faster R-CNN Zhipeng Yan, Moyuan Huang, and Hao Jiang: Incorporating Uncertainty Into Deep Model Akshaya Purohit, Lenord Melvix, Xiaoyu Zhou, and Bolun Zhang: Stereo Visual Odometry for Self Driving Cars Yuting Wang and Yuwei Wang: Video Captioning |
||
| June 7 |
Project presentations Anthony Thomas and Joni De Guzman: Parameterizing the Trifocal Tensor Adithya Seshasayee and Andrew Durnford: Stratified Autocalibration of Cameras using Convex Optimization Justin Gorgen, Haifeng Huang, Hao-en Sung, and Yen-Ting Chen: A Fused Convolutional and Recurrent Neural Network for Visual-Inertial Odometry Ishan Gupta and Shashank Tyagi: Face Analysis using Multi-task Learning Mike Liu, Jing Wang, and Yuhang Ming: Reimplementation of Struck: Structured Output Tracking with Kernels Harshita Mangal, Zhengqin Li, and Ji Dai: Unpaired Image-to-Image Translation using Cycle-consistent Generative Adversarial Network: a Review |
||
Projects:
| Group members | Project |
|---|---|
| Justin Gorgen, Haifeng Huang, Hao-en Sung, and Yen-Ting Chen | A Fused Convolutional and Recurrent Neural Network for Visual-Inertial Odometry |
| Mike Liu, Jing Wang, and Yuhang Ming | Reimplementation of Struck: Structured Output Tracking with Kernels |
| Anthony Thomas and Joni De Guzman | Parameterizing the Trifocal Tensor |
| Kwonjoon Lee, Nimish Srivastava, Songting Xu, and Ali Mirzaei | A Comparative Study of Different GANs |
| Ishan Gupta and Shashank Tyagi | Face Analysis using Multi-task Learning |
| Adithya Seshasayee and Andrew Durnford | Stratified Autocalibration of Cameras using Convex Optimization |
| Yuting Wang and Yuwei Wang | Video Captioning |
| Akshaya Purohit, Lenord Melvix, Xiaoyu Zhou, and Bolun Zhang | Stereo Visual Odometry for Self Driving Cars |
| Mingdong Wang and Yixin Yang | Object Detection for Self Driving using Faster R-CNN |
| Zhipeng Yan, Moyuan Huang, and Hao Jiang | Incorporating Uncertainty Into Deep Model |
| Harshita Mangal, Zhengqin Li, and Ji Dai | Unpaired Image-to-Image Translation using Cycle-consistent Generative Adversarial Network: a Review |
| Dhanesh Pradhan | Human Activity Recognition using Skeleton Data from RGBD Sensors |
Last update: June 3, 2017