Marigold V2 Hits New SOTA in Monocular Depth Estimation with Single-Step Diffusion Transformers
AntonObukhov1 · x · 2026-09-11
Marigold V2 (arXiv:2609.08084) repurposes diffusion transformers (DiT) into state-of-the-art monocular depth estimators, setting a new SOTA on Sintel surface normal estimation.
Key techniques:
- Single-step inference from pretrained multi-step flow-matching models, with optional quantization — preserving capacity while staying cheap to run
- Two fixes for naive-training artifacts: aligning internal representations with semantic features from ground truth, and a two-stage fine-tuning protocol built on a novel Sinkhorn-based loss
- Sharper, cleaner depth maps with strong out-of-distribution generalization; 16-26% AbsRel improvement over the previous best on KITTI and ETH3D
- Resolves fur, foliage, and hair-thin edges that eluded prior models
Authors are from EPFL, University of Bologna and others; evals are live on Papers with Code.
Related event: Marigold V2 Achieves SOTA Monocular Depth Estimation in a Single Step(2 posts)→
More from Multimodal
- Single image to walkable 3D: video flythrough turned into full gaussian splat — XRarchitect · 2026-10-31
- Spatial Hub aggregates ~5000 public COLMAP-compatible spatial datasets for world models — janusch_patas · 2026-10-31
- Single image to walkable 3D Gaussian splat scene using World Labs Atlas and WebGPU — XRarchitect · 2026-10-31
- 60+ community-made ChatGPT slash-style words turn one word into a full visual style — heyshrutimishra · 2026-10-31
- Cartesia's cascaded voice agents top the Speech Agent Arena, beating native speech-to-speech — rohanpaul_ai · 2026-10-31
- Hyperflow H3 hits #5 on I2V and #3 on video editing at under half Seedance 2.5's price — ArtificialAnlys · 2026-10-31