runyiyang/paperreading

📖 Paper reading notes in computer vision and machine learning, especially 3D SLAM, Implicit Representation and semantic segmantation. (constantly updating!!!) Everyone is welcomed to share your ideas and comments, I would greatly appreciate it if you can point out some of my mistakes or answer my questions.

This is a catalogue of my paper reading notes

​ There are mainly 3 parts of my reading.

​ The first part is mainly about Computer Vision where I'm concerned in most, I will try to read some classic paper intensively while do some paper reading extensively.

​ The second part of my reading is something about Machine Learning. It's an enormous topic. From that part I will read some classic and essential paper like NERF and Attention is all you need, to extend my vision and follow the routine.

​ The third part is going to be about Robotics/Dynamics/Control Theory .etc. I plan to only do some thesis study extensively in this part.

​ "*" means I think it's idea is brilliant.

Computer Vision

General Conclusion

3D SLAM

• ICCV 2015 *PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization
• DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation
• Learning Multi-Scene Absolute Pose Regression with Transformers
• CVPR2022(Oral) *RPMG: Projective Manifold Gradient Layer for Deep Rotation Regression
• PoseNetV2: Geometric loss functions for camera pose regression with deep learning
• Visual Odometry Revisited What Should Be Learnt
• Visual Camera Re-Localizationfrom RGB and RGB-D Images Using DSAC

Semantic Segmentation

• CVPR2022 ST++: Make Self-training Work Better for Semi-supervised Semantic Segmentation
• CVPR2022 Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels

Machine Learning

Robotics