mlsanityData to 7 Oct 2026

Towards Robust Driving Perception: A Flexible Scale-Driven Family for Self-Supervised Monocular Depth Estimation

arXiv 2607.00736 ↗First posted 1 Jul 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Self-Supervised Monocular Depth Estimation (MDE) has garnered attention in recent years due to its independence from ground truth. However, most existing models are limited to a single scale and exhibit considerable performance degradation in complex driving environments. Networks specifically designed to handle dynamic traffic participants tend to be overly complex, hindering their deployment on resource-constrained automotive edge devices. To address these limitations and move towards robust driving perception, we propose FlexDepth, a scale-driven and flexible family of self-supervised MDE models tailored for challenging road scenarios. FlexDepth employs a two-stage static-dynamic decoupled training strategy, enabling the independent assessment of confidence for both static backgrounds and dynamic road objects. Furthermore, it introduces a meticulously designed Scale-Driven Decoder (SDD) to dynamically select components based on scale size, facilitating efficient feature fusion and the output of high-precision depth maps. Extensive experiments on standard driving benchmarks demonstrate that without any auxiliary information, our model achieves state-of-the-art performance across arbitrary scales with minimal computational overhead. Our smallest model, Flex-Nano, requires only 0.7 GFLOPs and achieves 37.6 FPS on mobile platforms, ensuring reliable real-time perception while maintaining excellent zero-shot generalization. Our source code is available: https://github.com/startnew/flexdepth

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 43 of its 58 references that are in the map. First posted on arXiv 1 Jul 2026.

Builds on

43 of 58 references.

References · 43

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  4. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  5. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  6. 2021Vision Transformers for Dense Predictioncited by 1,789
  7. 2024Depth Anything V2cited by 1,385
  8. 2024Depth Anything: Unleashing the Power of Large-Scale Unlabeled Datacited by 1,346
  9. 2014Depth Map Prediction from a Single Image using a Multi-Scale Deep Networkcited by 1,264
  10. 2016Unsupervised Monocular Depth Estimation with Left-Right Consistencycited by 819
  11. 2017Unsupervised Learning of Depth and Ego-Motion from Videocited by 778
  12. 2014Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecturecited by 483
31 more
  1. 20193D Packing for Self-Supervised Monocular Depth Estimationcited by 437
  2. 2016Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescuecited by 368
  3. 2023Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Imagecited by 357
  4. 2017Sparsity Invariant CNNscited by 325
  5. 2021The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depthcited by 190
  6. 2018Unsupervised Learning of Monocular Depth Estimation and Visual Odometry with Deep Feature Reconstructioncited by 159
  7. 2017Learning Depth from Monocular Videos using Direct Methodscited by 154
  8. 2020Feature-metric Loss for Self-supervised Learning of Depth and Egomotioncited by 140
  9. 2022MonoViT: Self-Supervised Monocular Depth Estimation with a Vision Transformercited by 136
  10. 2019Self-Supervised Monocular Depth Hintscited by 127
  11. 2020Self-supervised Monocular Trained Depth Estimation using Self-attention and Discrete Disparity Volumecited by 116
  12. 2020HR-Depth: High Resolution Self-Supervised Monocular Depth Estimationcited by 115
  13. 2022Lite-Mono: A Lightweight CNN and Transformer Architecture for Self-Supervised Monocular Depth Estimationcited by 100
  14. 2020Unsupervised Monocular Depth Learning in Dynamic Scenescited by 84
  15. 2025Marigold: Affordable Adaptation of Diffusion-Based Image Generators for Image Analysiscited by 78
  16. 2021Learning Monocular Depth in Dynamic Scenes via Instance-Aware Projection Consistencycited by 58
  17. 2023Learning to Upsample by Learning to Samplecited by 57
  18. 2020Forget About the LiDAR: Self-Supervised Depth Estimators with MED Probability Volumescited by 50
  19. 2022Disentangling Object Motion and Occlusion for Unsupervised Multi-frame Monocular Depthcited by 48
  20. 2021StructDepth: Leveraging the structural regularities for self-supervised indoor depth estimationcited by 36
  21. 2023RM-Depth: Unsupervised Learning of Recurrent Monocular Depth in Dynamic Scenescited by 22
  22. 2023Dynamo-Depth: Fixing Unsupervised Depth Estimation for Dynamical Scenescited by 20
  23. 2025Multi-view Reconstruction via SfM-guided Monocular Depth Estimationcited by 17
  24. 2023DualRefine: Self-Supervised Depth and Pose Estimation Through Iterative Epipolar Sampling and Refinement Toward Equilibriumcited by 16
  25. 2023From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Priorcited by 13
  26. 2024ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusioncited by 6
  27. 2025Hybrid-grained Feature Aggregation with Coarse-to-fine Language Guidance for Self-supervised Monocular Depth Estimationcited by 6
  28. 2021Unsupervised Monocular Depth Perception: Focusing on Moving Objectscited by 4
  29. 2024Mining Supervision for Dynamic Regions in Self-Supervised Monocular Depth Estimationcited by 3
  30. 2024NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-trainingcited by 2
  31. 2026PuriLight: A Lightweight Shuffle and Purification Framework for Monocular Depth Estimationcited by 1

Built on by

Not cited yet.