mlsanityData to 7 Oct 2026

SS3D: End2End Self-Supervised 3D from Web Videos

arXiv 2604.22686 ↗First posted 24 Apr 2026cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

We present SS3D, a web-scale SfM-based self-supervision pretraining pipeline for feed-forward 3D estimation from monocular video. Our model jointly predicts depth, ego-motion, and intrinsics in a single forward pass and is trained/evaluated as a coherent end-to-end 3D estimator. To stabilize joint learning, we use an intrinsics-first two-stage schedule and a unified single-checkpoint evaluation protocol. Scaling SfM self-supervision to unconstrained web video is challenging due to weak multi-view observability and strong corpus heterogeneity; we address these with a multi-view signal proxy (MVS) used for filtering and curriculum sampling, and with expert training distilled into a single student. Pretraining on YouTube-8M (~100M frames after filtering) yields strong cross-domain zero-shot transfer and improved fine-tuning performance over prior self-supervised baselines. We release the pretrained checkpoint and code.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 62 of its 77 references that are in the map. First posted on arXiv 24 Apr 2026.

Builds on

62 of 77 references.

References · 62

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  5. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  6. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  7. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  8. 2021Vision Transformers for Dense Predictioncited by 1,789
  9. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
  10. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,484
  11. 2025VGGT: Visual Geometry Grounded Transformercited by 1,475
  12. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,346
38 more
  1. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,264
  2. 2018Multi-Task Learning as Multi-Objective Optimizationcited by 809
  3. 2017Unsupervised Learning of Depth and Ego-Motion from Videocited by 778
  4. 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningcited by 738
  5. 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 673
  6. 2023ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depthcited by 646
  7. 2025Depth Anything 3: Recovering the Visual Space from Any Viewscited by 626
  8. 2020AdaBins: Depth Estimation using Adaptive Binscited by 596
  9. 2025Continuous 3D Perception Model with Persistent Statecited by 527
  10. 20193D Packing for Self-Supervised Monocular Depth Estimationcited by 437
  11. 2024UniDepth: Universal Monocular Metric Depth Estimationcited by 375
  12. 2018GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Posecited by 341
  13. 2016YouTube-8M: A Large-Scale Video Classification Benchmarkcited by 324
  14. 2025MapAnything: Universal Feed-Forward Metric 3D Reconstructioncited by 295
  15. 2019Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Videocited by 244
  16. 2018Competitive Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentationcited by 222
  17. 2021The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depthcited by 190
  18. 2020On the uncertainty of self-supervised monocular depth estimationcited by 182
  19. 2023MiDaS v3.1 -- A Model Zoo for Robust Monocular Relative Depth Estimationcited by 173
  20. 2019Depth from Videos in the Wild: Unsupervised Monocular Depth Learning from Unknown Camerascited by 170
  21. 2018Unsupervised Learning of Monocular Depth Estimation and Visual Odometry with Deep Feature Reconstructioncited by 159
  22. 2018DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistencycited by 148
  23. 2020Self-Supervised Monocular Depth Estimation: Solving the Dynamic Object Problem by Semantic Guidancecited by 148
  24. 2020Feature-metric Loss for Self-supervised Learning of Depth and Egomotioncited by 140
  25. 2022MonoViT: Self-Supervised Monocular Depth Estimation with a Vision Transformercited by 136
  26. 2020Semantically-Guided Representation Learning for Self-Supervised Monocular Depthcited by 134
  27. 2019Self-supervised Learning with Geometric Constraints in Monocular Video: Connecting Flow, Depth, and Cameracited by 130
  28. 2025TTT3R: 3D Reconstruction as Test-Time Trainingcited by 122
  29. 2020HR-Depth: High Resolution Self-Supervised Monocular Depth Estimationcited by 115
  30. 2025VGGT-Long: Chunk it, Loop it, Align it -- Pushing VGGT's Limits on Kilometer-scale Long RGB Sequencescited by 107
  31. 2020Unsupervised Monocular Depth Learning in Dynamic Scenescited by 84
  32. 2020The Edge of Depth: Explicit Constraints between Segmentation and Depthcited by 72
  33. 2024Deep Patch Visual SLAMcited by 68
  34. 2021Self-Supervised Monocular Depth Estimation with Internal Feature Fusioncited by 67
  35. 2021Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection Consistencycited by 58
  36. 2026InfiniteVGGT: Visual Geometry Grounded Transformer for Endless Streamscited by 51
  37. 2022EfficientTrain: Exploring Generalized Curriculum Learning for Training Visual Backbonescited by 42
  38. 2021StructDepth: Leveraging the structural regularities for self-supervised indoor depth estimationcited by 36

50 most cited of 62 All 62 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026VLRC: Vision-Language Reprojection Consistency as a scalable signal for better feed-forward 3D pretrainingcited by 0