HyperAI超神经

首页算力平台文档资讯论文教程数据集百科 SOTA LLM 模型天梯 GPU 天梯顶会

中文

HyperAI超神经

Cross Domain Few Shot Object Detection On

评估指标

mAP

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
CD-ViTO	60.5	Cross-Domain Few-Shot Object Detection via Enhanced Open-Set Object Detector
BIOT(10-Shot)	58.4	Balanced ID-OOD tradeoff transfer makes query based detectors good few shot learners	-
BIOT(5-Shot)	53.3	Balanced ID-OOD tradeoff transfer makes query based detectors good few shot learners	-
DE-ViT-FT	49.2	Detect Everything with Few Examples
ViTDeT-FT	23.4	Exploring Plain Vision Transformer Backbones for Object Detection
FSCE	15.9	FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding
DeFRCN	15.5	DeFRCN: Decoupled Faster R-CNN for Few-Shot Object Detection
TFA w/cos	14.8	Frustratingly Simple Few-Shot Object Detection
Meta-RCNN	14.0	Meta-RCNN: Meta Learning for Few-Shot Object Detection	-
Detic-FT	12.0	Detecting Twenty-thousand Classes using Image-level Supervision

0 of 10 row(s) selected.

Cross Domain Few Shot Object Detection On | SOTA | HyperAI超神经