arXiv AI
GeoBridge: Evolving Transferable Decoder Topologies for Geometric Vision
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Computer Vision
Oct 7
DensiTok: Making Feed-Forward 3D Gaussian Splatting See More Views Than It Is Given
arXiv:2610.07958v1 Announce Type: new Abstract: Feed-forward 3D Gaussian Splatting (3DGS) reconstructs a scene in a single forward pass, replacing per-scene optimization with a network trained across...
arXiv AI
Oct 5
Less Decoder is More Encoder: Geometric Representation Learning from Novel View Synthesis
arXiv:2610.03717v1 Announce Type: cross Abstract: This paper examines the role of Novel View Synthesis (NVS) in geometric representation learning. In principle, NVS should reason about 3D scene struc...
arXiv AI
Jul 8
Harnessing Generative Image Models for Training-Free Primitive Shape Abstraction
arXiv:2607. 05568v1 Announce Type: cross Abstract: Representing 3D shapes as compact sets of geometric primitives is fundamental to robotics, simulation, and scene understanding.
arXiv Computer Vision
Oct 21
HERMES++: Toward a Unified Driving World Model for 3D Scene Understanding and Generation
HERMES++ is a unified driving world model that simultaneously performs 3D scene understanding and future geometry prediction. It uses a BEV representation to merge multi‑view spatial data for LLM compatibility, introduces LLM‑enhanced world queries for knowledge transfer, and employs a Current‑to‑Future Link to condition future geometry on semantic context. A Joint Geometric Optimization strategy enforces structural integrity by combining explicit geometric constraints with implicit latent regularization, and extensive benchmark tests show that HERMES++ outperforms specialist methods in both point cloud prediction and scene understanding.
arXiv Machine Learning
Oct 9
One Block, Multiple Depths: Recurrent Vision Transformers with Depth-Programmed Experts
The paper introduces reViT, a recurrent vision transformer that uses a single Transformer block repeatedly to achieve the accuracy of a full-depth encoder while keeping inference FLOPs comparable. It achieves depth‑specific transformations by expressing each block’s feed‑forward network as a convex combination of a small shared expert bank, with a continuous depth coordinate guiding the mixture. Experiments on ImageNet‑1k and DINOv2 distillation show that reViT matches or surpasses state‑of‑the‑art models, reduces stored parameters by about 70%, and supports elastic‑depth deployment from a single checkpoint.
arXiv Computer Vision
Oct 5
GeoScaffold: Learning Compact Geometric Latents via Reconstruction for Efficient Vision-Language Navigation
GeoScaffold introduces a geometric supervision framework that embeds 3D geometry into vision‑language navigation policies during training, eliminating the need for depth sensors or 3D encoders at inference. The method learns a compact depth tokenizer from training trajectories, then fine‑tunes the policy with geometry query tokens that reconstruct depth, connectivity, and traversability, turning these states into compact geometric latents for action decoding. After training, the tokenizer, generators, and reconstruction heads are discarded, leaving a lightweight backbone that outperforms vision‑only navigators on continuous VLN benchmarks.