| Vision as Unified Multimodal Generation | ✓ Link | 0.059 | 0.959 | SenseNova-Vision | 2026-07-07 |
| Lotus-2: Advancing Geometric Dense Prediction with Powerful Image Generative Model | ✓ Link | 0.067 | 0.945 | Lotus-2 (zero-shot) | 2025-11-30 |
| DINOv3 | ✓ Link | 0.073 | 0.967 | DINOv3 + Depth Anything V2 | 2025-08-13 |
| Depth Anything V2 | ✓ Link | 0.074 | | Depth Anything V2 (ViT-L) | 2024-06-13 |
| Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data | ✓ Link | 0.076 | 0.947 | Depth Anything (ViT-L) | 2024-01-19 |
| CH3Depth: Efficient and Flexible Depth Foundation Model with Flow Matching | | 0.076 | 0.949 | CH3Depth (96K) | 2025-06-11 |
| DepthART: Scaling Foundation Monocular Depth to Tiny Models | ✓ Link | 0.079 | 0.933 | DepthART-L (zero-shot) | 2026-07-19 |
| Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Prediction | ✓ Link | 0.081 | 0.931 | Lotus-D (zero-shot) | 2024-09-26 |
| DepthART: Scaling Foundation Monocular Depth to Tiny Models | ✓ Link | 0.082 | 0.930 | DepthART-S (zero-shot) | 2026-07-19 |
| Video Generative Models as Geometry Learner | ✓ Link | 0.082 | 0.926 | GeoNeXt | 2026-08-28 |
| AnyDepth: Depth Estimation Made Easy | ✓ Link | 0.086 | 0.926 | AnyDepth ViT-L | 2026-01-06 |
| Let RGB Be the Language of Vision | ✓ Link | 0.093 | 0.901 | RINO-Zero (Qwen-Image-Edit) | 2026-07-14 |
| ZipDepth: Bringing Lightweight Zero-Shot Monocular Depth Anywhere, on Any Device | ✓ Link | 0.123 | 0.864 | ZipDepth | 2026-07-09 |
| Depth Anything V2 | ✓ Link | | 0.948 | Depth Anything V2 (ViT-G) | 2024-06-13 |
| Depth Anything 3: Recovering the Visual Space from Any Views | ✓ Link | | 0.971 | Depth Anything 3 Mono | 2025-11-13 |
| In Pursuit of Pixel Supervision for Visual Pre-training | ✓ Link | | 0.934 | Pixio-H + DPT | 2025-12-17 |