CS 7931 Computer Vision Seminar - Spring 2019

Instructor: Srikumar Ramalingam

Lecture Time: Tue 10:45PM-11:45 AM

Place: MEB 3485

Abstract:

This course focuses on recent trends in computer vision especially the deep learning algorithms such as CNNs, dropouts, residual networks, deep learning representations, semantic segmentation, stereo reconstruction, image-based localization, etc. The presentations will be given by the instructor, invited speakers (at least 2 to 3 from industrial research labs), and students.

Class Schedule

Date	Lecture Number	Topic
1/8	L1	Srikumar Ramalingam on "Linear Regions and Expressiveness of Deep Neural Networks"
1/15	L2	Xin Yu on “VLASE: Localization using Semantic Boundaries”
1/22	L3	Siddhant Ranade on “Novel Constraints For Single View Manhattan 3D Reconstruction”
1/29	L4	Abhinav Kumar on “License Plate Re-identification using Neural Embedding of Fisher Vectors”
2/5	L5	Remaldeep Singh on “Low latency semantic video segmentation” from CVPR 2018
2/12	L6	Wenzheng Tao “Geometric deep learning on graphs and manifolds using mixture model CNNs” from CVPR 2017
2/19	L7	Mohammad Rehan Ghori on “Image-to-Image Translation with Conditional Adversarial Networks”
2/26	L8	cGAN continued …
3/5	L9	Qingkai Lu on “Mask R-CNN”
3/12		Spring Break
3/19	L10	Nicholas Harrison on FlowNet: Learning Optical Flow with Convolutional Networks
3/26	L11	Rui Jin on LeafSnap
4/9	L12	Xin Yu (Intro to Pytorch)
4/16	L13	Srikumar Ramalingam (Novel applications of Submodularity)
4/26	L14	On Friday: Seminar by Alyosha Efros from UC Berkeley on GANs (WEB 3780)

Tentative Reading List for students

• Learning Hierarchical Features for Scene Labeling, Clement Farabet, Camille Couprie, Laurent Najman and Yann LeCun, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013.

• Representation Learning: A Review and New Perspectives, Yoshua Bengio, Aaron Courville, Pascal Vincent, Arxiv, 2012.

• ImageNet Classification with Deep Convolutional Neural Networks, Alex Krizhevsky, Ilya Sutskever, Geoffrey E Hinton, NIPS 2012.

• Goodfellow, I., Warde-Farley, D., Mirza, M., Courville, A., and Bengio, Y. (2013). Maxout networks. Technical Report, Universite de Montreal.

• https://www.cs.toronto.edu/~hinton/absps/JMLRdropout.pdf

• Jure Zbontar and Yann LeCun, Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches, 2016.

• David Eigen, Christian Puhrsch, and Rob Fergus, Depth Map Prediction from a Single Image using a Multi-Scale Deep Network, NIPS, 2014.

Honor Policy

Students are expected to work on their own, as instructed by the Professor. Students may discuss concepts with other individuals either in the class or outside the class, but they may not receive results electronically from any source that is not documented in their report. Any student who is found to be violating this policy will be given a failing grade for the course and will be reported to the authorities as described in the University's Student Code.

Accommodations Policy

The University of Utah seeks to provide equal access to its programs, services and activities for people with disabilities. If you will need accommodations in the class, reasonable prior notice needs to be given to the Center for Disability Services, 162 Olpin Union Building, 581-5020 (V/TDD). CDS will work with you and the instructor to make arrangements for accommodations. All written information in this course can be made available in alternative format with prior notification to the Center for Disability Services.

Grading

• Class Presentation and participation: 100%