Notice
Recent Posts
Recent Comments
Link
| 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 |
| 31 |
Tags
- 인공지능
- convnext
- NLP
- Swin Transformer
- Convolution Block Attention Module
- GAN # Label Smoothing
- Transformer
- multi head attention
- DeepLearning
- simclr
- Semantic Segmentation
- ECCV
- contrastive learning
- Deepfake
- You Only Look Once #YOLO
- Instance segmentation
- 딥러닝
- BERT
- Attention
- FLOPS
- boundary unlearning
- ConvNet
- mordernize
- class activation mapping
- machine unlearning
- CBAM
- NT-Xent(InFoNCE)
- self supervised learning
- Bidirectional Transformers for Language Understanding
- ICML
Archives
- Today
- Total
목록Instance segmentation (1)
륵늘
Segmentation 정리
CNN 모델에서 할 수 있는 Task Classification - 분류 Object Detection - 물체 탐지 Segmentation - 픽셀별 분류 Comparision about Semantic, Instance and Panoptic Semantic은 나무, 하늘, 사람, 자동차 등과 같은 것들과 배경을 포함해서 pixel 별로 class를 구별 Instance는 배경을 제외한 나무, 하늘 사람, 자동차 등과 같은 것을 각각을 구별 ex) 사람1, 사람2, 사람3, 자동차1, 자동차2, 자동차3 Panoptic = Semantic + Instance Segmantic Segmentation 픽셀이 어느 class에 속하는지 예측하는 것. 지금 그림에서는 class가 5개이고, 각각의 픽셀들에..
딥러닝/VISION
2023. 7. 26. 20:01