Papers with Code · CC BY-SA 4.0

NYU-Depth V2

아카이브 평가 85건 · 마지막 적재/변경 2026-10-30

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
HybridDepth	RMSE 0.128 absolute relative error 0.026 Delta < 1.25 0.988 Delta < 1.25^2 1.000 Delta < 1.25^3 1.000	HybridDepth: Robust Metric Depth Fusion by Leveraging Depth from Focus and Single-Image Priors 2024-07-26	코드 1
Distill Any Depth	absolute relative error 0.043 Delta < 1.25 0.981	Distill Any Depth: Distillation Creates a Stronger Monocular Depth Estimator 2025-02-26	코드 1
UniK3D (FT, metric)	RMSE 0.173 absolute relative error 0.044 Delta < 1.25 0.989 Delta < 1.25^2 0.998 Delta < 1.25^3 1.000 log 10 0.019	UniK3D: Universal Camera Monocular 3D Estimation 2025-03-20	코드 1
UniDepthV2 (FT, metric)	RMSE 0.180 absolute relative error 0.046 Delta < 1.25 0.988 Delta < 1.25^2 0.998 Delta < 1.25^3 1.000 log 10 0.020	UniDepthV2: Universal Monocular Metric Depth Estimation Made Simpler 2025-02-27	코드 1
PrimeDepth + Depth Anything	absolute relative error 0.046 Delta < 1.25 0.977	PrimeDepth: Efficient Monocular Depth Estimation with a Stable Diffusion Preimage 2024-09-13	코드 1
Metric3Dv2(L, FT)	RMSE 0.183 absolute relative error 0.047 Delta < 1.25 0.989 Delta < 1.25^2 0.998 Delta < 1.25^3 1.000 log 10 0.020	Metric3Dv2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimation 2024-03-22	코드 1
DepthMaster	absolute relative error 0.050 Delta < 1.25 0.972	DepthMaster: Taming Diffusion Models for Monocular Depth Estimation 2025-01-05	코드 1
GRIN	RMSE 0.251 absolute relative error 0.051	GRIN: Zero-Shot Metric Depth with Pixel-Level Diffusion 2024-09-15	연결 없음
Marigold + E2E FT(zero-shot)	absolute relative error 0.052 Delta < 1.25 0.966	Fine-Tuning Image-Conditional Diffusion Models is Easier than You Think 2024-09-17	코드 1
Marigold	RMSE 0.224 absolute relative error 0.055 Delta < 1.25 0.964 Delta < 1.25^2 0.991 Delta < 1.25^3 0.998 log 10 0.024	Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation 2023-12-04	코드 1 · 코드 2 · 코드 3 · 코드 4
Depth Anything	RMSE 0.206 absolute relative error 0.056 Delta < 1.25 0.984 Delta < 1.25^2 0.998 Delta < 1.25^3 1.000 log 10 0.024	Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data 2024-01-19	코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5 · 코드 6 · 코드 7
UniDepth (Zero-shot)	RMSE 0.201 absolute relative error 0.058 Delta < 1.25 0.984 Delta < 1.25^2 0.997 Delta < 1.25^3 0.999 log 10 0.024	UniDepth: Universal Monocular Metric Depth Estimation 2024-03-27	코드 1 · 코드 2 · 코드 3
PrimeDepth	absolute relative error 0.058 Delta < 1.25 0.966	PrimeDepth: Efficient Monocular Depth Estimation with a Stable Diffusion Preimage 2024-09-13	코드 1
ECoDepth	RMSE 0.218 absolute relative error 0.059 Delta < 1.25 0.978 Delta < 1.25^2 0.997 Delta < 1.25^3 0.999 log 10 0.026	ECoDepth: Effective Conditioning of Diffusion Models for Monocular Depth Estimation 2024-03-27	코드 1
MetaPrompt-SD	RMSE 0.223 absolute relative error 0.061 Delta < 1.25 0.976 Delta < 1.25^2 0.997 Delta < 1.25^3 0.999 log 10 0.027	Harnessing Diffusion Models for Visual Perception with Meta Prompts 2023-12-22	코드 1
EVP	RMSE 0.224 absolute relative error 0.061 Delta < 1.25 0.976 Delta < 1.25^2 0.997 Delta < 1.25^3 0.999 log 10 0.027	EVP: Enhanced Visual Perception using Inverse Multi-Attentive Feature Refinement and Regularized Image-Text Alignment 2023-12-13	코드 1
TADP	RMSE 0.225 absolute relative error 0.062 Delta < 1.25 0.976 Delta < 1.25^2 0.997 Delta < 1.25^3 0.999 log 10 0.027	Text-image Alignment for Diffusion-based Perception 2023-09-29	코드 1 · 코드 2
FutureDepth	RMSE 0.233 absolute relative error 0.063 Delta < 1.25 0.981 Delta < 1.25^2 0.996 Delta < 1.25^3 0.999 log 10 0.027	FutureDepth: Learning to Predict the Future Improves Video Depth Estimation 2024-03-19	연결 없음
MeSa	RMSE 0.238 absolute relative error 0.066 Delta < 1.25 0.964 Delta < 1.25^2 0.995 Delta < 1.25^3 0.999 log 10 0.029	MeSa: Masked, Geometric, and Supervised Pre-training for Monocular Depth Estimation 2023-10-06	연결 없음
PolyMaX(ConvNeXt-L)	RMSE 0.25 absolute relative error 0.067 Delta < 1.25 0.969 Delta < 1.25^2 0.9958 Delta < 1.25^3 0.999 log 10 0.029	PolyMaX: General Dense Prediction with Mask Transformer 2023-11-09	코드 1
VPD	RMSE 0.254 absolute relative error 0.069 Delta < 1.25 0.964 Delta < 1.25^2 0.995 Delta < 1.25^3 0.999 log 10 0.030	Unleashing Text-to-Image Diffusion Models for Visual Perception 2023-03-03	코드 1 · 코드 2
NVDS(DPT-L)	RMSE 0.282 absolute relative error 0.072 Delta < 1.25 0.9493 Delta < 1.25^2 0.991 Delta < 1.25^3 0.997 log 10 0.031	NVDS+: Towards Efficient and Versatile Neural Stabilizer for Video Depth Estimation 2023-07-17	코드 1 · 코드 2
DMD	RMSE 0.296 absolute relative error 0.072 Delta < 1.25 0.953 Delta < 1.25^2 0.989 Delta < 1.25^3 0.996 log 10 0.031	Zero-Shot Metric Depth with a Field-of-View Conditioned Diffusion Model 2023-12-20	연결 없음
ScaleDepth-N	RMSE 0.267 absolute relative error 0.074 Delta < 1.25 0.957 Delta < 1.25^2 0.994 Delta < 1.25^3 0.999 log 10 0.032	ScaleDepth: Decomposing Metric Depth Estimation into Scale Prediction and Relative Depth Estimation 2024-07-11	코드 1
DepthGen	RMSE 0.314 absolute relative error 0.074 Delta < 1.25 0.946 Delta < 1.25^2 0.987 Delta < 1.25^3 0.996 log 10 0.032	Monocular Depth Estimation using Diffusion Models 2023-02-28	연결 없음
ZoeD-M12-N	RMSE 0.270 absolute relative error 0.075 Delta < 1.25 0.955 Delta < 1.25^2 0.995 Delta < 1.25^3 0.999 log 10 0.032	ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth 2023-02-23	코드 1 · 코드 2 · 코드 3 · 코드 4 · 코드 5 · 코드 6
AiT-P(SwinV2-L)	RMSE 0.275 absolute relative error 0.076 Delta < 1.25 0.954 Delta < 1.25^2 0.994 Delta < 1.25^3 0.999 log 10 0.033	All in Tokens: Unifying Output Space of Visual Tasks via Soft Token 2023-01-05	코드 1
Gaming for Depth (GfD)	RMSE 0.364 absolute relative error 0.080 Delta < 1.25 0.931 Delta < 1.25^2 0.986 Delta < 1.25^3 0.996 log 10 0.033	Large-scale Monocular Depth Estimation in the Wild 2023-09-18	연결 없음
SwinV2-L 1K-MIM	RMSE 0.287 absolute relative error 0.083 Delta < 1.25 0.949 Delta < 1.25^2 0.994 Delta < 1.25^3 0.999 log 10 0.035	Revealing the Dark Secrets of Masked Image Modeling 2022-05-26	코드 1
Metric3D (ConvNeXt-Large, Zero-shot testing)	RMSE 0.310 absolute relative error 0.083 Delta < 1.25 0.944 Delta < 1.25^2 0.986 Delta < 1.25^3 0.995 log 10 0.035	Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Image 2023-07-20	코드 1