Research map / 3D vision / Depth and stereo
ZipDepth: Bringing Lightweight Zero-Shot Monocular Depth Anywhere, on Any Device
Fabio Tosi, Luca Bartolomei, Matteo Poggi, Stefano Mattoccia
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add SE and Global Context attention in Stage 3 (ablation) | ZipDepth · NYUv2 | AbsRel: 8.7 -> 8.4 | Improved |
| add SE and Global Context attention in Stage 3 (ablation) | ZipDepth · KITTI | AbsRel: 12.6 -> 12.3 | Improved |
| add SPPF and cross-scale refinement (ablation) | ZipDepth · KITTI | AbsRel: 13.1 -> 12.3 | Improved |
| add SPPF and cross-scale refinement (ablation) | ZipDepth · NYUv2 | AbsRel: 9.0 -> 8.4 | Improved |
| add half-resolution skip path (ablation) | ZipDepth · NYUv2 | AbsRel: 9.1 -> 8.4 | Improved |
| add half-resolution skip path (ablation) | ZipDepth · DIODE | AbsRel: 23.8 -> 22.6 | Improved |
| add half-resolution skip path (ablation) | ZipDepth · Jetson Orin NX 15W | FPS: 48.9 -> 34.4 | Hurt |
| use convex upsampling (GPU path) instead of bilinear (ablation) | ZipDepth · UnrealStereo4K | SI-BF1: 0.088 -> 0.120 | Improved |
| use convex upsampling (NPU path) instead of bilinear (ablation) | ZipDepth · UnrealStereo4K | SI-BF1: 0.088 -> 0.105 | Improved |
| use convex upsampling instead of bilinear (ablation) | ZipDepth · NYUv2 | AbsRel: 8.3 -> 8.4 | No clear effect |
| use GPU-path convex upsampling instead of NPU path (ablation) | ZipDepth · iPhone 12 ANE | FPS: 375 -> 240 | Hurt |
| increase distillation data from 1% to 100% of 14.1M images (ablation) | ZipDepth · NYUv2 | AbsRel: 12.2 -> 8.4 | Improved |
Closest papers by meaning
- UniDepth: Universal Monocular Metric Depth Estimation (CVPR 2024 · similarity 0.89)
- BetterDepth: Plug-and-Play Diffusion Refiner for Zero-Shot Monocular Depth Estimation (NeurIPS 2024 · similarity 0.89)
- Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data (CVPR 2024 · similarity 0.88)
- SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation (CVPR 2025 · similarity 0.88)
- DCDepth: Progressive Monocular Depth Estimation in Discrete Cosine Domain (NeurIPS 2024 · similarity 0.88)
- Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
