Codesota · Benchmark · NYU Depth V2Home/Leaderboards/Vision & Documents/Depth Estimation/NYU Depth V2
Unknown

NYU Depth V2.

NYU Depth V2 is a state-of-the-art machine learning benchmark indexed on Codesota. This page tracks published model results, top scores per metric, and the SOTA timeline for NYU Depth V2.

Paper ↗Leaderboard ↓
§ 01 · SOTA history

Year over year.

§ 02 · Leaderboard

Results by metric.

Found a wrong score or missing run?
Use row edits to send a sourced correction into moderation.
Add / edit result ↗Report issue ↗

Absrel

Absrel is the reported evaluation metric for NYU Depth V2. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Absrelverifiedpapervendorcommunityunverified

Muted rows were not state of the art when published — an earlier or same-year result already scored better.

RankModelTrustScoreYearLinksFix
01ZoeDepth-N
ZoeDepth-N (metric depth estimation). NYU Depth V2 test split. Table 1 in paper.
verified0.072023Source ↗Looks wrong?
02Marigold
Marigold (diffusion-based monocular depth). Affine-invariant evaluation on NYU Depth V2. AbsRel reported in Table 1.
verified0.062023Source ↗Looks wrong?
03MiDaS 3.1 (BEiT-512)
MiDaS v3.1 BEiT-512 model. AbsRel on NYU Depth V2 test set. Reported in Depth Anything Table 1.
verified0.052024Source ↗Looks wrong?
04Depth Anything V1 (ViT-L)
Depth Anything V1 ViT-L, fine-tuned on NYU Depth V2. AbsRel on test set. Table 1.
verified0.042024Source ↗Looks wrong?
05Depth Anything V2 (ViT-L)
Depth Anything V2 ViT-L, fine-tuned metric model. AbsRel on NYU Depth V2 test set. Table 2 in paper.
verified0.042024Source ↗Looks wrong?
§ 04 · Submit a result

Add to the leaderboard.

← Back to Depth Estimation