2d-to-3d feature distillation 13d shape understanding 1feedforward networks 1mesh analysis 1unsupervised 3d learning 1
From the 1 of 3 linked papers with an AI index.
3 papers
cs.CV2026
MeshFM: 2D Features Are All You Need for 3D Shape Understanding
Jinfan Zhou, Richard Liu, Itai Lang +1
MeshFM is a feedforward framework that learns 3D shape features by distilling 2D features from visual foundation models using a two‑stage training process that requires no 3D annot…
cs.CV2025
Generative Spatiotemporal Data Augmentation
Jinfan Zhou, Lixin Luo, Sungmin Eum +2
We explore spatiotemporal data augmentation using video foundation models to diversify both camera viewpoints and scene dynamics. Unlike existing approaches based on simple geometr…
cs.CV2025
Taming Feed-forward Reconstruction Models as Latent Encoders for 3D Generative Models
Suttisak Wizadwongsa, Jinfan Zhou, Edward Li +1
Recent AI-based 3D content creation has largely evolved along two paths: feed-forward image-to-3D reconstruction approaches and 3D generative models trained with 2D or 3D supervisi…