My repository: https://github.com/Chen-Yulin/Semantic-SAM
My venv: ssam

Overview of the proposed SceneGraphFusion framework. Our method takes a stream of RGB-D images a) as input to create an incremental geometric segmentation b). Then, the properties of each segment and a neighbor graph between segments are constructed. The properties d) and neighbor graph e) of the segments that have been updated in the current frame c) are used as the inputs to compute node and edge features f) and to predict a 3D scene graph g). Finally, the predictions are h) fused back into a globally consistent 3D graph.

这片文章可以成为场景物理重建的基石之一
类似的后续工作有OMG-Seg

注：此DINO并非自蒸馏自监督的那个[[DINO]]，而是派生自[[DETR]]

A vision-language model that unifies vision-language understanding and generation tasks.

将不同帧$X_t$中的特征集合在M中特征点的公式：