Survey benchmarks SSL instance discrimination and masked image modeling for object detection, finding instance discrimination suits CNN encoders while MIM suits ViT encoders and custom pre-training, especially for small objects.
Momentum contrast for unsupervised visual representation learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2024 2verdicts
UNVERDICTED 2representative citing papers
ReCLIP++ rectifies class and space biases in CLIP via separate reference and positional features, logit subtraction, and a mask decoder with contrastive loss to improve unsupervised semantic segmentation on PASCAL VOC, ADE20K and other benchmarks.
citing papers explorer
-
Self-Supervised Learning for Real-World Object Detection: a Survey
Survey benchmarks SSL instance discrimination and masked image modeling for object detection, finding instance discrimination suits CNN encoders while MIM suits ViT encoders and custom pre-training, especially for small objects.
-
ReCLIP++: Learn to Rectify the Bias of CLIP for Unsupervised Semantic Segmentation
ReCLIP++ rectifies class and space biases in CLIP via separate reference and positional features, logit subtraction, and a mask decoder with contrastive loss to improve unsupervised semantic segmentation on PASCAL VOC, ADE20K and other benchmarks.