Get 3D models of multiple objects in RGB video with RayTran
Get 3D models of multiple objects in RGB video with RayTran
RayTran: 3D pose estimation and shape reconstruction of multiple objects from videos with ray-traced transformers
arXiv paper abstract https://arxiv.org/abs/2203.13296
arXiv PDF paper https://arxiv.org/pdf/2203.13296.pdf
... propose a transformer-based neural network architecture for multi-object 3D reconstruction from RGB videos.
... represent its knowledge: as a global 3D grid of features and an array of view-specific 2D grids. ... exchange information between the two with a dedicated bidirectional attention mechanism.
... exploit knowledge about the image formation ... to ... sparsify the attention weight matrix, making ... feasible on current hardware, both in terms of memory and computation.
... attach a DETR-style head ... to detect the objects in the scene and to predict their 3D pose and 3D shape.
... architecture is single stage, end-to-end trainable, and it can reason holistically about a scene from multiple video frames
... method ... outperform (1) recent state-of-the-art methods for 3D object pose estimation from RGB videos ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments