| 일 | 월 | 화 | 수 | 목 | 금 | 토 |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | |||
| 5 | 6 | 7 | 8 | 9 | 10 | 11 |
| 12 | 13 | 14 | 15 | 16 | 17 | 18 |
| 19 | 20 | 21 | 22 | 23 | 24 | 25 |
| 26 | 27 | 28 | 29 | 30 |
- Diffusion
- mobilenet
- Focal loss
- Unreal Engine
- Image denoising
- convolution
- Vision Transformer
- CNN
- posec3d
- DDPM
- 티스토리챌린지
- Deep Learning
- 모수 추정 방법
- computer graphics
- depthwise convolution
- Swin Transformer
- 오블완
- 최대 사후 확률
- covolution
- deformable covolution
- 최대 가능도 추정
- Action Recognition
- 포인터
- C++
- OpenGL
- 딥러닝
- Today
- Total
목록AI Project (2)
심드렁하게 저장
개요PoseC3D는 2021년 CVPR Workshop에서 발표된 논문에서 제안된 모델로 논문의 풀 네임은 "Revisiting Skeleton-based Action Recognition"이다. Skeleton Based Action Recognition 모델로서 2D Keypoint를 입력으로 받아 사람의 행동을 인식한다. 해당모델은 꽤 오래전 논문이지만 아직까지도 Papers with code 기준 상위권에 위치하고 있다. 또한 훈련 dataset을 구성할때 2D Keypoint를 사용한다는 점에서 활용 난이도도 높지 않을것으로 예상되어 행동 인식 프로젝트의 메인 모델로 채택하게 되었다. 1. GOAL - 본 논문의 목적 본 논문 이전 기존 Skeleton-based Action Recogniti..
RuntimeError: CUDA error: device-side assert triggeredCUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.For debugging consider passing CUDA_LAUNCH_BLOCKING=1.Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.이 에러는 GPU에서 수행 중인 연산에서 잘못된 인덱스/값이 들어가서 assert가 실패했을 때 발생 대표적인 원인잘못된 라벨 인덱스예: CrossEntropyLoss 사용 중인데 ta..