SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.04554 · CVPR · 2022

ChiTransformer: Towards Reliable Stereo from Cues

Qing Su, Shihao Ji

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

Current stereo matching techniques are challenged by restricted searching space, occluded regions and sheer size. While single image depth estimation is spared from these challenges and can achieve satisfactory results with the extracted monocular cues, the lack of stereoscopic relationship renders the monocular prediction less reliable on its own especially in highly dynamic or cluttered environments. To address these issues in both scenarios, we present an optic-chiasm-inspired self-supervised binocular depth estimation method, wherein vision transformer (ViT) with a gated positional cross-attention (GPCA) layer is designed to enable feature-sensitive pattern retrieval between views, while retaining the extensive context information aggregated through self-attentions. Monocular cues from a single view are thereafter conditionally rectified by a blending layer with the retrieved pattern pairs. This crossover design is biologically analogous to the optic-chasma structure in human visual system and hence the name, Chi-Transformer. Our experiments show that this architecture yields substantial improvements over state-of-the-art selfsupervised stereo approaches by 11%, and can be used on both rectilinear and non-rectilinear (e.g., fisheye) images.

For agents

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

get_harvested_code_for_paper("2203.04554")
get_code_for_paper("2203.04554")
have("2203.04554")

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

curl https://syntology.ai/api/ran/2203.04554.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.04554.svg)](https://syntology.ai/paper/2203.04554)

Connect an agent — have() is free.