SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.17638 · CVPR · 2024

Learning with Unreliability: Fast Few-shot Voxel Radiance Fields with Relative Geometric Consistency

Hao Tang, Shengfeng He, Bangzhen Liu, Bailin Deng, Yingjie Xu

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

a) Input Views (b) Warped Images (c) w/o Unreliability (d) Ours (e) Comparisons on the Realistic Synthetic 360 • dataset [24] Figure 1. We present ReVoRF, a voxel-based framework designed to capitalize on the unreliability inherent in warped novel views. (b) demonstrates the warping outcomes, where black holes signify unmatched pixels from the original view. (c) illustrates the results of training when these holes are masked out, which unfortunately results in ambiguous geometric structures. In contrast, (d) showcases our approach's ability to maintain correct geometric consistency. ReVoRF achieves this by leveraging relational depth prior knowledge within these unreliable hole regions. Our approach demonstrates the best reconstruction quality while being one of the fastest few-shot approaches in (e).

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2403.17638")
get_code_for_paper("2403.17638")
have("2403.17638")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2403.17638.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2403.17638.svg)](https://syntology.ai/paper/2403.17638)

Connect an agent — have() is free.