Current 3D scene segmentation methods are heavily dependent on manually annotated 3D
training datasets. Such manual annotations are labor-intensive, and often lack fine-grained …

In this work, we address the task of point cloud denoising using a novel framework adapting
Diffusion Schrödinger bridges to unstructured data like point sets. Unlike previous works that …

While recent foundational video generators produce visually rich output, they still struggle
with appearance drift, where objects gradually degrade or change inconsistently across …

We present DeSiRe-GS, a self-supervised gaussian splatting representation, enabling
effective static-dynamic decomposition and high-fidelity surface reconstruction in complex …

Given that visual foundation models (VFMs) are trained on extensive datasets but often
limited to 2D images, a natural question arises: how well do they understand the 3D world …

Large-scale pre-trained vision models are becoming increasingly prevalent, offering
expressive and generalizable visual representations that benefit various downstream tasks …

Vision foundation models, particularly the ViT family, have revolutionized image
understanding by providing rich semantic features. However, despite their success in 2D …

Inspired by the emergent 3D capabilities in image generators, we explore whether video
generators similarly exhibit 3D awareness. Using structure-from-motion (SfM) as a …

Emerging 3D geometric foundation models, such as DUSt3R, offer a promising approach for
in-the-wild 3D vision tasks. However, due to the high-dimensional nature of the problem …

Diffusion-based video generators are now a reality. Being trained on a large corpus of real
videos, such models can generate diverse yet realistic videos (Brooks et al., 2024; Zheng et …