An insanely technically challenging project has come to a glorious end! I can't emphasize the amount of hard work behind this effort which I hope will influence the trajectory of 3D perception
SAM 3D wins a CVPR 2026 Best Paper Honorable Mention for extracting human-object interaction trajectories from video
The framework generates 3D content from standard images.
Users congratulated the Meta SAM 3D team on its CVPR 2026 honorable mention and voiced excitement about applying the work to robotics pipelines.
Most Activity
SAM 3D: 3Dfy Anything in Images received an honorable mention for best paper at CVPR 2026. Even more satisfying, it has enabled learning human-object-interaction trajectories from video for training robots (I am at ICRA and had numerous conversations on this!). You can read the paper at https://openaccess.thecvf.com/content/CVPR2026/html/Chen_SAM_3D_3Dfy_Anything_in_Images_CVPR_2026_paper.html
An insanely technically challenging project has come to a glorious end! I can't emphasize the amount of hard work behind this effort which I hope will influence the trajectory of 3D perception
@JitendraMalikCV Congratulations @JitendraMalikCV and Sam 3D team!!
SAM 3D: 3Dfy Anything in Images received an honorable mention for best paper at CVPR 2026. Even more satisfying, it has enabled learning human-object-interaction trajectories from video for training robots (I am at ICRA and had numerous conversations on this!). You can read the paper at https://openaccess.thecvf.com/content/CVPR2026/html/Chen_SAM_3D_3Dfy_Anything_in_Images_CVPR_2026_paper.html

@JitendraMalikCV Congratulations on the honorable mention for SAM 3D! The potential to learn human-object-interaction trajectories for robotics is incredibly exciting. Well-deserved recognition at CVPR 2026.
@JitendraMalikCV Congratulations Jitendra! We look forward on using it in our pipelines!
SAM 3D: 3Dfy Anything in Images received an honorable mention for best paper at CVPR 2026. Even more satisfying, it has enabled learning human-object-interaction trajectories from video for training robots (I am at ICRA and had numerous conversations on this!). You can read the paper at https://openaccess.thecvf.com/content/CVPR2026/html/Chen_SAM_3D_3Dfy_Anything_in_Images_CVPR_2026_paper.html