Research map / 3D vision / Depth and stereo
Kick Back & Relax: Learning to Reconstruct the World by Watching SlowTV
Jaime Spencer, Chris Russell, Simon Hadfield, Richard Bowden
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 |
|---|---|---|---|
| train self-supervised depth on the SlowTV dataset plus Mannequin Challenge and Kitti instead of automotive data only | SS-MDE (ConvNeXt-B) · zero-shot outdoor and indoor datasets | depth accuracy (AbsRel, delta, F-score) | Improved |
| leave-one-out removal of aspect ratio augmentation, learned intrinsics, random support frames or flexible pose (ablation) | full SlowTV model · Kitti, Mannequin Challenge, SYNS | depth accuracy | Improved |
| learn camera intrinsics with the pose network instead of using ground-truth (ablation) | full SlowTV model · Kitti (calibrated) | depth accuracy | No clear effect |
| learn camera intrinsics when ground truth intrinsics come from COLMAP (ablation) | full SlowTV model · Mannequin Challenge | depth accuracy | Improved |
| add minimum reconstruction loss and automasking from Monodepth2 (ablation) | monocular depth training · subset of three training datasets | depth accuracy | Improved |
| median alignment instead of scale-and-shift alignment | evaluation · all datasets | depth accuracy | Hurt |
| self-supervised SlowTV model versus supervised SOTA on indoor datasets | SlowTV model · indoor datasets | depth accuracy | Hurt |
| SlowTV-trained model on Map-free relocalization versus self-supervised baselines | SlowTV model · MapFreeReloc val | relocalization errors | Improved |
Closest papers by meaning
- Seurat: From Moving Points to Depth (CVPR 2025 · similarity 0.87)
- From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Prior (CVPR 2024 · similarity 0.86)
- Video Depth Anything: Consistent Depth Estimation for Super-Long Videos (CVPR 2025 · similarity 0.86)
- Synthetic-to-Real Self-supervised Robust Depth Estimation via Learning with Motion and Structure Priors (CVPR 2025 · similarity 0.86)
- Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data (CVPR 2024 · similarity 0.86)
- One View Is Enough: In-the-Wild Monocular Pretraining for Novel View Generation (NeurIPS 2026 · similarity 0.86)
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.
