AI Paper Atlas

Research map / 3D vision / Depth and stereo

Monocular Depth Estimation Using Multi-Scale Continuous CRFs as Sequential Deep Networks

Dan Xu, Elisa Ricci, Wanli Ouyang, Xiaogang Wang, Nicu Sebe

TPAMI 2018 · 3D vision › Depth and stereo · approach: Dense depth prediction and completion · 110 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add multi-scale cascade CRFs on top of the front-end (ablation)ResNet-50 · NYU Depth V2rel error: 0.168 -> 0.143Improved
add multi-scale cascade CRFs on top of the front-end (ablation)VGG-CD · NYU Depth V2rel error: 0.203 -> 0.169Improved
use deeper front-end (ResNet-50 instead of AlexNet) (ablation)pretrained front-end · NYU Depth V2rel error: 0.265 -> 0.168Improved
use symmetric encoder-decoder (VGG-CD instead of VGG-16) (ablation)pretrained front-end · NYU Depth V2rel error: 0.228 -> 0.203Improved
add end-to-end CRF fusion instead of post-processing C-CRF (ablation)VGG-CD · NYU Depth V2rel error: 0.193 -> 0.187Improved
add continuous CRF post-processing (ablation)VGG-CD · NYU Depth V2rel error: 0.203 -> 0.193Improved
use CRF-based fusion instead of HED weighted averaging (ablation)VGG-CD · NYU Depth V2rel error: 0.185 -> 0.169Improved
increase number of scales from 1 to 5 (cascade) (ablation)cascade CRF · NYU Depth V2rel error: 0.187 -> 0.169Improved
use unified model instead of cascade (5-scale) (ablation)VGG-CD · NYU Depth V2rel error: 0.169 -> 0.163Improved
use all-to-one message passing instead of top-down (ablation)cascade CRF ResNet-50 · NYU Depth V2rel error: 0.175 -> 0.154Improved
use skip-connection instead of bottom-up message passing (ablation)cascade CRF ResNet-50 · NYU Depth V2rel error: 0.169 -> 0.161Improved
add 95K image pretraining data (vs 4.7K)ResNet-50-unified bottom up · NYU Depth V2rel error: 0.139 -> 0.121Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy