-
How to write good papers
- Good Citizen of CVPR at CVPR 2018 in Salt Lake City, Utah
- How to write a great research paper by Microsoft Research
-
Application
- Face Recognition
[1][2][6] - Face Super-Resolution
[11] - Image Caption
[3] - Person Re-identification
[4][7][10][12][14] - Object Detection
[5] - Image/Instance Segmentation
[8][13]
- Face Recognition
-
Method
- Loss Design
[1][5] - Attention
[3][13] - GAN
[4][7][8][9][14] - Domain Adaption
[4][6][7][9] - Clustering
[12]
- Loss Design
[1] ECCV'18: Orthogonal Deep Features Decomposition for Age-Invariant Face Recognition | Tencent AI
[2] ECCV'18: GridFace: Face Rectification via Learning Local Homography Transformations | Face++ | Resource
[3] CVPR'17: SCA-CNN: Spatial and Channel-wise Attention in Convolutional Networks for Image Captioning | NUS | Resource | Code
[4] ECCV'18: Generalizing A Person Retrieval Model Hetero- and Homogeneously | ANU | Code
[5] CVPR'18: Repulsion Loss: Detecting Pedestrians in a Crowd | Face++ | Resource | Code
[6] Arxiv'18: DocFace: Matching ID Document Photos to Selfies | MSU | Code
[7] CVPR'18: Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification | ANU | Code
[8] ACM MM'18: Understanding Humans in Crowded Scenes: Deep Nested Adversarial Learning and A New Benchmark for Multi-Human Parsing | NUS | Resource | Code
[9] NIPS'18: A Unified Feature Disentangler for Multi-Domain Image Translation and Manipulation | NTU, Taiwan | Code
[10] ECCV'18: Person Search via A Mask-Guided Two-Stream CNN Model | NUST | Resource
[11] CVPR'18: FSRNet: End-to-End Learning Face Super-Resolution with Facial Priors | NUST | Resource | Code
[12] AAAA'19: A Bottom-up Clustering Approach to Unsupervised Person Re-identification | UTS
[13] CVPR'16: Attention to Scale: Scale-aware Semantic Image Segmentation | UCLA, Baidu | Code
[14] NIPS'18: FD-GAN: Pose-guided Feature Distilling GAN for Robust Person Re-identification | CUHK, SenseTime | Code
Loss:
CVPR'19: SoDeep: A Sorting Deep Net to Learn Ranking Loss Surrogates
Continues Learning:
CVPR'19: Learning to Remember: A Synaptic Plasticity Driven Framework for Continual Learning
Noise Label:
CVPR'19: Weakly Supervised Image Classification Through Noise Regularization
CVPR'19: Noise-Tolerant Paradigm for Training Face Recognition CNNs
CVPR'19: On Stabilizing Generative Adversarial Training With Noise
CVPR'19: Probabilistic End-To-End Noise Correction for Learning With Noisy Labels
CVPR'19: MetaCleaner: Learning to Hallucinate Clean Representations for Noisy-Labeled Visual Recognition
Unbalance Data:
CVPR'19: Learning Not to Learn: Training Deep Neural Networks With Biased Data
CVPR'19: Class-Balanced Loss Based on Effective Number of Samples
Domain Adaptation:
CVPR'19: Unsupervised Domain Adaptation Using Feature-Whitening and Consensus Loss
CVPR'19: Sliced Wasserstein Generative Models
CVPR'19: Domain-Specific Batch Normalization for Unsupervised Domain Adaptation
Self-supervised:
CVPR'19: Self-Supervised GANs via Auxiliary Rotation Loss
CVPR'19: Self-Supervised Representation Learning by Rotation Feature Decoupling
memeda:
CVPR'19: Rethinking the Evaluation of Video Summaries