Research map / 3D vision / Depth and stereo
Depth AnyEvent: A Cross-Modal Distillation Paradigm for Event-Based Monocular Depth Estimation
Luca Bartolomei, Enrico Mannocci, Fabio Tosi, 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 cross-modal distillation from DAv2 proxy labels instead of synthetic-only training | E2Depth · MVSEC | Abs Rel: 0.527 -> 0.400 | Improved |
| add cross-modal distillation instead of synthetic-only training | EReFormer · MVSEC | Abs Rel: 0.518 -> 0.448 | Improved |
| add cross-modal distillation instead of synthetic-only training | DepthAnyEvent · DSEC | Abs Rel: 0.297 -> 0.213 | Improved |
| use distillation instead of ground-truth supervision | DepthAnyEvent-R · DSEC | Abs Rel: 0.191 -> 0.226 | Hurt |
| use distillation instead of ground-truth supervision | EReFormer · MVSEC | Abs Rel: 0.511 -> 0.448 | Improved |
| use distillation instead of ground-truth supervision | E2Depth · MVSEC | Abs Rel: 0.420 -> 0.400 | Improved |
| add recurrent module (DepthAnyEvent-R) to DepthAnyEvent (ablation) | DepthAnyEvent · DSEC | Abs Rel: 0.201 -> 0.191 | Improved |
| replace DAv2 with Depth Pro as distillation teacher (ablation) | DepthAnyEvent-R · MVSEC | Abs Rel: 0.399 -> 0.429 | Hurt |
| use Tencode instead of Voxel Grid representation (ablation) | DepthAnyEvent-R · MVSEC | Abs Rel: 0.382 -> 0.365 | Improved |
| add DAv2 pretraining (ablation) | DepthAnyEvent-R · MVSEC | Abs Rel: 0.446 -> 0.365 | Improved |
| combine distillation with ground-truth labels (ablation) | DepthAnyEvent-R · MVSEC | Abs Rel: 0.365 -> 0.362 | Improved |
| use pretrained VFM backbone instead of E2Depth | E2Depth · DSEC (zero-shot) | Abs Rel: 0.395 -> 0.276 | Improved |
Later papers that compared with this method
1 later accepted paper reports Depth AnyEvent as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- EventVGGT: Exploring Cross-Modal Distillation for Consistent Event-based Depth Estimation (ECCV 2026 · better on 4 of 4 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Depth Any Event Stream: Enhancing Event-based Monocular Depth Estimation via Dense-to-Sparse Distillation (ICCV 2025 · similarity 0.94)
- EventVGGT: Exploring Cross-Modal Distillation for Consistent Event-based Depth Estimation (ECCV 2026 · similarity 0.94)
- Event-Driven Dynamic Scene Depth Completion (NeurIPS 2025 · similarity 0.90)
- DERD-Net: Learning Depth from Event-based Ray Densities (NeurIPS 2025 · similarity 0.88)
- Zero-Shot Event-Intensity Asymmetric Stereo via Visual Prompting from Image Domain (NeurIPS 2024 · similarity 0.87)
- Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data (CVPR 2024 · 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.
