Advanced Topics in Computer Vision

(2019 Graduate Course)

Instructor: Prof. Mu Yadong (email: myd@pku.edu.cn)
Location: Room 104, Building 3
Time: Tuesday 10:10am - 12:00pm (biweekly), Thursday 10:10am - 12:00pm (weekly)
TA: n/a
Office Hour: Friday 10am-12pm
Feb 21, 2019 Introduction - I
  • Course Introduction
  • Introduction to Computer Vision
Feb 26, 2019 Introduction - II
  • Basics of Deep Neural Networks
Feb 28, 2019 Topic: Visual Recognition - I
  • Visual Recognition: Task Definition and Challenges
  • Bag-of-words Models
  • Spatial Pyramid Matching & Pyramid Match Kernel
  • Vocabulary Tree, Sparse Coding
March 7, 2019 No Class  
March 12, 2019 Topic: Visual Recognition - II
  • Deep Learning for Visual Recognition: LeNet-5, AlexNet, VGG-16, GoogleNet, ResNet, DenseNet, DualPathNet
March 14, 2019 Topic: Object Detection - I
  • V-J Face Detector (Integral Image, AdaBoost, Cascade)
  • HOG+SVM with NMS
March 21, 2019 Topic: Object Detection - II
  • Deformable Part Model (DPM) for Pedestrian Detection
  • R-CNN
  • Fast R-CNN
March 26, 2019 Topic: Object Detection - II
  • Faster R-CNN
  • R-FCN
  • Multi-Scale R-CNN
  • Feature Pyramid Network
  Topic: Pixel Computing - I
  • Pixel Labeling: Segmentation, Matting, Parsing
  • Unsupervised Image Segmentation: K-means, Mean-Shift, Normalized Cut
March 28, 2019 Topic: Pixel Computing - II
  • Interactive Object Cutout: GraphCut, GrabCut, LazySnapping
  • Image Matting: Poisson Matting, Closed-Form Matting, Robust Color Sampling
  • Image Co-segmentation
  • Image Inpainting / Image Completion
April 4, 2019 Topic: Pixel Computing - III
  • Deep Pixel Labeling: FCN, DeepLab, SegNet, CNN-as-RNN
  • Human Pose Estimation: Bottom-Up and Top-Down
April 9, 2019 No class  
April 11, 2019 Topic: Large-Scale Image Search
  • Dimension Reducation: PCA, CCA, Fisher LDA
  • Nonlinear Methods: MDS, ISOMAP, LLE
  • LPP, graph embedding
  • Johnson Lindenstrauss lemma
  • The magic of hashing collision: Bloom Filter
April 18, 2019 Paper Presentation - I
  • Lin Zhongya, Pyramid Stereo Matching Network, CVPR 2018
  • Shi Xiangyu, A Coarse-to-fine Pyramidal Model for Person Re-identification via Multi-Loss dynamic training, CVPR 2019
  • Jin Jiandong, Vehicle Re-Identification with the Space-Time Prior, CVPR 2018
  • Liu Kun, Learning Transferable Architectures for Scalable Image Recognition, CVPR 2018
  • Zheng Liangfeng, Looking for the Devil in the Details: Learning Trilinear Attention Sampling Network for Fine-grained Image Recognition, https://arxiv.org/abs/1903.06150
April 23, 2019 Paper Presentation - II
  • Zhang Ziqi, Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attacks, CVPR 2019
  • Gong Guoqiang, Dynamic Zoom-in Network for Fast Object Detection in Large Images, CVPR 2018
  • Guan Yushuo, DSFD: Dual Shot Face Detector, CVPR 2019
  • Zhao Hongye, AON: Towards Arbitrarily Oriented Text Recognition, CVPR 2018
  • Guo Dewen, S. Iizuka, E. Simo-Serra, and H. Ishikawa. Globally and locally consistent image completion. ACM Transactions on Graphics (TOG), 36(4):107, 2017.
April 25, 2019 Paper Presentation - III
  • Zhuang Nan, Multi-Label Image Recognition with Graph Convolutional Networks, CVPR 2019
  • Su Shupeng, Collective Matrix Factorization Hashing for Multimodal Data, CVPR 2014 / Unsupervised Deep Hashing via Binary Latent Factor Models for Large-scale Cross-modal Retrieval, IJCAI 2018
  • Li Yongzhi, Thoracic Disease Identification and Localization with Limited Supervision, CVPR 2018
  • Wang Shen, Look More Than Once: An Accurate Detector for Text of Arbitrary Shapes, CVPR 2019
  • Lv Guannan, Simple Black-Box Adversarial Perturbations for Deep Networks, https://arxiv.org/abs/1612.06299
May 7, 2019 Topic: Large-Scale Image Search
  • Locality-Sensitive Hashing: the concept and proof of sublinear complexity in the STOC98 paper
  • LSH schemes for Hamming space, cosine similarity, Jaccard index (minHash)
  • Spectral Hashing
  • ITQ
  • Semi-Supervsied Hashing
  • Supervised Hashing with Kernel
  • Deep Hashing
  • Discrete Hashing
  • Hashing for Large-Scale Machine Learning
May 9, 2019 Topic: Video Computing
  • Introduction of Video Computing Tasks
  • Video Features (STIP, Deep Video, C3D, Trajectory Feature)
  • Deep Learning for Video Classification (multi-stream fusion techniques)
  • An Illustrative System for Video Classification
  • TRECVID Tasks (MED and Instance Search)
May 16, 2019 Topic: Reccurent Deep Networks
  • Unrolling Computational Graph
  • RNN variants (recurrent through output, sequence-input-single-output, teaching forcing, encoder-decoder, bi/quad-directional RNN etc.)
  • Generative RNN modeling
  • Back propagation through time (BPTT)
  • The issue and remedy for long-term dependency in RNN
  • Long short-term memory (LSTM)
  • Applications (image captioning, convLSTM for rainfall prediction, social LSTM)
May 21, 2019 Topic: Autonomous Vehicle
  • Past, Present, and Prospect
  • Funding, Challenges, and Benchmarks
  • DeepLanes
  • CCF-UISEE Traffic Sign Detection
  • End-to-End Driving Model
  • Deep Driving
  • Localization by Visual Odometry
  • HMM based Driver Maneuver Prediction
May 24, 2019 Invited Speech
  • Two speakers from ByteDance and Megvii
May 30, 2019 Course Project Presentation - I  
June 4, 2019 Course Project Presentation - II  
June 6, 2019 Course Project Presentation - III