https://github.com/yunqing-me/attackvlm
Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
Human Activity Recognition is the problem of identifying events performed by humans given a video input. It is formulated as a binary (or multiclass) classification problem of outputting activity class labels. Activity Recognition is an important problem with many societal applications including smart surveillance, video search/retrieval, intelligent robots, and other monitoring systems.
Source: Learning Latent Sub-events in Activity Videos Using Temporal Attention Filters
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks
Learning Transferable Visual Models From Natural Language Supervision
Masked Feature Prediction for Self-Supervised Visual Pre-Training
Temporal Convolutional Networks for Action Segmentation and Detection
Benchmarks and Challenges in Pose Estimation for Egocentric Hand Interactions with Objects
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
Learning Transferable Visual Models From Natural Language Supervision
Temporal Localization of Fine-Grained Actions in Videos by Domain Transfer from Web Images
VideoLSTM Convolves, Attends and Flows for Action Recognition
Semi-Supervised Classification with Graph Convolutional Networks
Frozen Transformers in Language Models Are Effective Visual Encoder Layers
Moments in Time Dataset: one million videos for event understanding
Model Editing as a Robust and Denoised variant of DPO: A Case Study on Toxicity
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
You Only Watch Once: A Unified CNN Architecture for Real-Time Spatiotemporal Action Localization
Blindly Assess Quality of In-the-Wild Videos via Quality-aware Pre-training and Motion Perception
BSN: Boundary Sensitive Network for Temporal Action Proposal Generation
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
DF-Net: The Digital Forensics Network for Image Forgery Detection
DF-Net: The Digital Forensics Network for Image Forgery Detection
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.