Maxime Oquab

Papers

Total Citations

H-Index

About

Maxime Oquab is a leading researcher in computer vision, with a primary focus on unsupervised learning, visual representation learning, and image animation. His most influential work, "Self-appearance-aided Differential Evolution for Motion Transfer," addresses the challenge of unsupervised motion transfer—animating a static source image using the motion from a driving video without requiring labeled data or domain-specific priors. This contribution is pivotal for advancing realistic image animation while preserving the identity of the source object. Although this specific paper has garnered 2 citations, Oquab’s broader impact is far more substantial, as he is widely recognized for his pioneering contributions to self-supervised learning, particularly through the development of DINO and DINOv2. These frameworks have revolutionized visual representation learning by enabling models to learn rich, semantic features without human annotations, achieving state-of-the-art performance on numerous benchmarks and amassing thousands of citations. Oquab’s work has been instrumental in making unsupervised learning practical and scalable, influencing both academic research and industrial applications in computer vision.