Signals & Systems:


1. 了解该领域的研究前沿,用latex+texbib写一份 Related works (at least One full Page) (2周)
2. 理解该论文,并实现 Interim Report (methods) (3周)
3. 理解该论文,并实现 DemoShow (results) (5周)
4. 针对该论文问题提出改进思路并实现 FinalReport&Code (5周)

A.Stereo Matching
A1. Computing the Stereo Matching Cost with a Convolutional Neural Network
A2. Slanted Plane Smoothing Stereo
A3. Object Scene Flow for Autonomous Vehicles
A4. A Multi-Block-Matching Approach for Stereo
A5. Piecewise Rigid Scene Flow
A6. Stereo Image Warping for Improved Depth Estimation of Road Surfaces
A7. Stereo Processing by Semiglobal Matching and Mutual Information
A8. A Two-Stage Correlation Method for Stereoscopic Depth Estimation

B. Detection
B1. DenseBox: Unifying Landmark Localization with End to End Object Detection
B2. Learning to Detect Vehicles by Clustering Appearance Patterns
B3. Integrating Context and Occlusion for Car Detection by Hierarchical And-Or Model.
B4. Fast and Robust Object Detection Using Visual Subcategories

C. Tracking
C1. Globally-Optimal Greedy Algorithms for Tracking a Variable Number of Objects
C2. Discrete-Continuous Optimization for Multi-Target Tracking
C3. Continuous Energy Minimization for Multitarget Tracking
C4. Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor
C5. Detection- and Trajectory-Level Exclusion in Multiple Object Tracking
C6. Continuous Energy Minimization for Multitarget Tracking

D. Convolutional Neural Network/LSTM/Metric Learning
D1. Fast R-CNN: Fast Region-based Convolutional networks for object detection
D2. Translating Videos to Natural Language Using Deep Recurrent Neural Networks
D3. Traffic Sign Detection based on convolutional neural networks
D4. Sequence to Sequence – Video to Text
D5. Long-term Recurrent Convolutional Networks for Visual Recognition and Description