HyperAI超神経

プラットフォーム

Depth Estimation On Nyu Depth V2

評価指標

RMS

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

		Paper Title
PAD-Net	0.792	PAD-Net: Multi-Tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing
MS-CRF	0.586	Multi-Scale Continuous CRFs as Sequential Deep Networks for Monocular Depth Estimation
DORN	0.509	Deep Ordinal Regression Network for Monocular Depth Estimation
Freeform	0.433	Deep Optics for Monocular Depth Estimation and 3D Object Detection
Optimized, freeform	0.4325	Deep Optics for Monocular Depth Estimation and 3D Object Detection
VNL	0.416	Enforcing geometric constraints of virtual normal for depth prediction
BTS	0.407	From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimation
TransDepth (AGD+ ViT)	0.365	Transformer-Based Attention Networks for Continuous Pixel-Wise Prediction
AdaBins	0.364	AdaBins: Depth Estimation using Adaptive Bins
P3Depth	0.356	P3Depth: Monocular Depth Estimation with a Piecewise Planarity Prior
SwinV2-B 1K-MIM	0.304	Revealing the Dark Secrets of Masked Image Modeling
Semantic-aware NN	0.30	3D Ken Burns Effect from a Single Image
SwinV2-L 1K-MIM	0.287	Revealing the Dark Secrets of Masked Image Modeling
DINOv2 (ViT-g/14 frozen, w/ DPT decoder)	0.279	DINOv2: Learning Robust Visual Features without Supervision
EVP	0.224	EVP: Enhanced Visual Perception using Inverse Multi-Attentive Feature Refinement and Regularized Image-Text Alignment
Defocus/DepthNet (Normalized)	-	Focus on defocus: bridging the synthetic to real domain gap for depth estimation
A2J	-	A2J: Anchor-to-Joint Regression Network for 3D Articulated Pose Estimation from a Single Depth Image

0 of 17 row(s) selected.

Depth Estimation On Nyu Depth V2 | SOTA | HyperAI超神経