반응형
250x250
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 |
Tags
- polling-based object probing evaluation
- dinov2: learning robust visual features without supervision 논문
- 엔트로피란
- 논문 요약
- clip adapter
- vlm 환각
- 딥러닝 목적함수
- mobilenetv1
- object detection
- 원격 학습 안끊기게
- 딥러닝 엔트로피
- vlm
- dinov2 논문 리뷰
- 논문 리뷰
- 이미지 필터링
- 기계학습
- blip-2
- Object detection article
- 1차 미분 마스크
- dinov2: learning robust visual features without supervision
- evaluating object hallucination in large vision-language models
- evaluating object hallucination in large vision-language models paper
- 에지 검출
- vlm hallucination
- vlm hallucination paper
- evaluating object hallucination in large vision-language models 논문
- vlm 환각이란
- 객체 검출
- clip
- dinov2: learning robust visual features without supervision 논문 리뷰
Archives
- Today
- Total
목록tiny object detection (1)
My Vision, Computer Vision

AI-TOD_ICPR_camera_ready.pdf drive.google.com 본 논문의 main contribution 항공 이미지의 소형 객체 탐지를 위한 AI-TOD dataset을 소개한다. 소형 객체 탐지를 위한 네트워크 제안(M-CenterNet) → 이 내용은 본 글에 담지 않았음. Abstract AI-TOD dataset의 개요는 아래와 같다. Image : 28,036개 Object : 700,621개 Class : 8개 객체 평균 크기 : 12.8픽셀(제일 큰 객체 64픽셀) Introduction Tiny object detecion(소형 객체 탐지)는 대규모 감시, 지능형 교통, 위치 기반 서비스 등에 이용된다. 기존 객체 검출 태스크에 비해 항공 이미지 내 객체가 미시적인 경..
Paper
2024. 3. 25. 14:08