A comprehensive survey on contrastive learning

H Hu, X Wang, Y Zhang, Q Chen, Q Guan - Neurocomputing, 2024‏ - Elsevier
Contrastive Learning is self-supervised representation learning by training a model to
differentiate between similar and dissimilar samples. It has been shown to be effective and …

Fake it till you make it: Learning transferable representations from synthetic imagenet clones

MB Sarıyıldız, K Alahari, D Larlus… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
Recent image generation models such as Stable Diffusion have exhibited an impressive
ability to generate fairly realistic images starting from a simple text prompt. Could such …

Backbone is all your need: A simplified architecture for visual object tracking

B Chen, P Li, L Bai, L Qiao, Q Shen, B Li, W Gan… - European conference on …, 2022‏ - Springer
Exploiting a general-purpose neural architecture to replace hand-wired designs or inductive
biases has recently drawn extensive interest. However, existing tracking approaches rely on …

Usb: A unified semi-supervised learning benchmark for classification

Y Wang, H Chen, Y Fan, W Sun… - Advances in …, 2022‏ - proceedings.neurips.cc
Semi-supervised learning (SSL) improves model generalization by leveraging massive
unlabeled data to augment limited labeled samples. However, currently, popular SSL …

Siamese image modeling for self-supervised vision representation learning

C Tao, X Zhu, W Su, G Huang, B Li… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
Self-supervised learning (SSL) has delivered superior performance on a variety of
downstream vision tasks. Two main-stream SSL frameworks have been proposed, ie …

A fractional gradient descent algorithm robust to the initial weights of multilayer perceptron

X **e, YF Pu, J Wang - Neural Networks, 2023‏ - Elsevier
For multilayer perceptron (MLP), the initial weights will significantly influence its
performance. Based on the enhanced fractional derivative extend from convex optimization …

Humanbench: Towards general human-centric perception with projector assisted pretraining

S Tang, C Chen, Q **e, M Chen… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
Human-centric perceptions include a variety of vision tasks, which have widespread
industrial applications, including surveillance, autonomous driving, and the metaverse. It is …

A foundation language-image model of the retina (flair): Encoding expert knowledge in text supervision

J Silva-Rodriguez, H Chakor, R Kobbi, J Dolz… - Medical Image …, 2025‏ - Elsevier
Foundation vision-language models are currently transforming computer vision, and are on
the rise in medical imaging fueled by their very promising generalization capabilities …

A new benchmark: On the utility of synthetic data with blender for bare supervised learning and downstream domain adaptation

H Tang, K Jia - Proceedings of the IEEE/CVF conference on …, 2023‏ - openaccess.thecvf.com
Deep learning in computer vision has achieved great success with the price of large-scale
labeled training data. However, exhaustive data annotation is impracticable for each task of …

The tunnel effect: Building data representations in deep neural networks

W Masarczyk, M Ostaszewski, E Imani… - Advances in …, 2023‏ - proceedings.neurips.cc
Deep neural networks are widely known for their remarkable effectiveness across various
tasks, with the consensus that deeper networks implicitly learn more complex data …