HyperAI초신경
홈뉴스최신 연구 논문튜토리얼데이터셋백과사전SOTALLM 모델GPU 랭킹컨퍼런스
전체 검색
소개
HyperAI초신경

Cross Domain Few Shot Object Detection On 4

평가 지표

mAP

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름
mAP
Paper TitleRepository
FSCE12.0FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding
DeFRCN12.1DeFRCN: Decoupled Faster R-CNN for Few-Shot Object Detection
Detic-FT16.8Detecting Twenty-thousand Classes using Image-level Supervision
TFA w/cos11.8Frustratingly Simple Few-Shot Object Detection
Meta-RCNN11.2Meta-RCNN: Meta Learning for Few-Shot Object Detection-
BIOT(5-shot)18.0Balanced ID-OOD tradeoff transfer makes query based detectors good few shot learners-
CD-ViTO7.0Cross-Domain Few-Shot Object Detection via Enhanced Open-Set Object Detector
ViTDeT-FT15.8Exploring Plain Vision Transformer Backbones for Object Detection
DE-ViT-FT5.4Detect Everything with Few Examples
BIOT(10-shot)20.4Balanced ID-OOD tradeoff transfer makes query based detectors good few shot learners-
0 of 10 row(s) selected.