cd /news/computer-vision/pxdepth-pixel-space-modeling-for-str… · home topics computer-vision article
[ARTICLE · art-102394] src=arxiv.org ↗ pub= topic=computer-vision verified=true sentiment=· neutral

PXDepth: Pixel-Space Modeling for Structure Preserving Monocular Depth Estimation

Researchers propose PXDepth, a monocular depth estimation model that separates global context modeling from pixel-level prediction to preserve fine-grained structures and object boundaries. The model uses a large-patch Vision Transformer for global context and Context-Modulated Pixel Transformer blocks for high-resolution spatial representations, achieving competitive zero-shot depth accuracy while maintaining efficiency. Code and models are available at the project page.

read1 min views1 publishedAug 19, 2026

arXiv:2608.16984v1 Announce Type: new Abstract: Recent monocular depth estimators achieve strong zero-shot generalization, yet often struggle to preserve fine-grained structures and object boundaries. We attribute this limitation to the prevalent combination of large-patch ViT encoders and convolutional decoders, as coarse tokenization can weaken pixel-level cues that upsampling cannot fully recover. To address this issue, we propose PXDepth, a discriminative monocular depth model that separates global context modeling from pixel-level depth prediction. Specifically, a large-patch ViT captures global scene context, while a pixel-space predictor composed of Context-Modulated Pixel Transformer blocks maintains high-resolution spatial representations throughout depth estimation. This design preserves fine structures and sharp boundaries without sacrificing global depth consistency. Across diverse zero-shot benchmarks, PXDepth combines faithful local geometry with competitive global depth accuracy while remaining efficient at inference. Our code and model are available at https://yuanzhy29.github.io/PXDepth-Page/.

── more in #computer-vision 4 stories · sorted by recency
── more on @pxdepth 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/pxdepth-pixel-space-…] indexed:0 read:1min 2026-08-19 ·