Maxim Tatarchenko
Papers
2
Total Citations
25
H-Index
2
About
Maxim Tatarchenko is a computer vision researcher whose work centers on 3D shape understanding and self-supervised learning for robotics. His major contributions lie in developing convolutional neural networks that can jointly predict 3D object shape and estimate viewpoint from a single 2D image—a notoriously difficult inverse problem. Tatarchenko’s key innovation is the use of self-supervision: his network learns solely from object silhouettes in the input image, eliminating the need for expensive ground-truth 3D annotations. This approach, detailed in his most-cited paper (23 citations), enables scalable training from readily available imagery and is directly applicable to robotic manipulation and scene understanding. By circumventing the data bottleneck, his work advances practical 3D perception systems that can operate in real-world, unconstrained environments. Tatarchenko’s research bridges the gap between geometric computer vision and embodied AI, offering a path toward robots that can infer full 3D structure from fleeting, monocular observations. His contributions are particularly notable for their elegance and efficiency, making 3D reasoning more accessible to the broader robotics community.
Research Focus
Key Achievements
Top Papers
- 1
- 2