# Multi-Frame Self-Supervised Depth with Transformers

arXiv: https://arxiv.org/abs/2204.07616 · HTML: https://mlsanity.com/paper/2204.07616/

## Key facts

- Cited by 64 papers in the map (data through 14 Oct 2026).
- Builds on 75 of its 95 references that are in the map.
- First posted on arXiv 15 Apr 2022.

## Builds on (most cited first, 30 of 75)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [pixelNeRF: Neural Radiance Fields from One or Few Images](https://mlsanity.com/paper/2012.02190/) (2020)
- [SuperGlue: Learning Feature Matching with Graph Neural Networks](https://mlsanity.com/paper/1911.11763/) (2019)
- [PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume](https://mlsanity.com/paper/1709.02371/) (2017)
- [A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation](https://mlsanity.com/paper/1512.02134/) (2015)
- [NeRF++: Analyzing and Improving Neural Radiance Fields](https://mlsanity.com/paper/2010.07492/) (2020)
- [Unsupervised Monocular Depth Estimation with Left-Right Consistency](https://mlsanity.com/paper/1609.03677/) (2016)
- [Efficient Transformers: A Survey](https://mlsanity.com/paper/2009.06732/) (2020)
- [Unsupervised Learning of Depth and Ego-Motion from Video](https://mlsanity.com/paper/1704.07813/) (2017)
- [Depth-supervised NeRF: Fewer Views and Faster Training for Free](https://mlsanity.com/paper/2107.02791/) (2021)
- [Deep Ordinal Regression Network for Monocular Depth Estimation](https://mlsanity.com/paper/1806.02446/) (2018)
- [DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras](https://mlsanity.com/paper/2108.10869/) (2021)
- [TartanAir: A Dataset to Push the Limits of Visual SLAM](https://mlsanity.com/paper/2003.14338/) (2020)
- [Pyramid Stereo Matching Network](https://mlsanity.com/paper/1803.08669/) (2018)
- [Tracking Objects as Points](https://mlsanity.com/paper/2004.01177/) (2020)
- [FCOS3D: Fully Convolutional One-Stage Monocular 3D Object Detection](https://mlsanity.com/paper/2104.10956/) (2021)
- [3D Packing for Self-Supervised Monocular Depth Estimation](https://mlsanity.com/paper/1905.02693/) (2019)
- [From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimation](https://mlsanity.com/paper/1907.10326/) (2019)
- [Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue](https://mlsanity.com/paper/1603.04992/) (2016)
- [End-to-End Learning of Geometry and Context for Deep Stereo Regression](https://mlsanity.com/paper/1703.04309/) (2017)
- [Virtual KITTI 2](https://mlsanity.com/paper/2001.10773/) (2020)

## Built on by (most cited first, 30 of 64)

- [Unifying Flow, Stereo and Depth Estimation](https://mlsanity.com/paper/2211.05783/) (2022)
- [Towards Zero-Shot Scale-Aware Monocular Depth Estimation](https://mlsanity.com/paper/2306.17253/) (2023)
- [Learning Temporally Consistent Video Depth from Video Diffusion Priors](https://mlsanity.com/paper/2406.01493/) (2024)
- [Depth Any Video with Scalable Synthetic Data](https://mlsanity.com/paper/2410.10815/) (2024)
- [STS: Surround-view Temporal Stereo for Multi-view 3D Detection](https://mlsanity.com/paper/2208.10145/) (2022)
- [R3D3: Dense 3D Reconstruction of Dynamic Scenes from Multiple Cameras](https://mlsanity.com/paper/2308.14713/) (2023)
- [Cross-View Completion Models are Zero-shot Correspondence Estimators](https://mlsanity.com/paper/2412.09072/) (2024)
- [Learning to Fuse Monocular and Multi-view Cues for Multi-frame Depth Estimation in Dynamic Scenes](https://mlsanity.com/paper/2304.08993/) (2023)
- [Deep Digging into the Generalization of Self-Supervised Monocular Depth Estimation](https://mlsanity.com/paper/2205.11083/) (2022)
- [Synthetic-to-Real Self-supervised Robust Depth Estimation via Learning with Motion and Structure Priors](https://mlsanity.com/paper/2503.20211/) (2025)
- [Jasmine: Harnessing Diffusion Prior for Self-supervised Depth Estimation](https://mlsanity.com/paper/2503.15905/) (2025)
- [MACARONS: Mapping And Coverage Anticipation with RGB Online Self-Supervision](https://mlsanity.com/paper/2303.03315/) (2023)
- [EGA-Depth: Efficient Guided Attention for Self-Supervised Multi-Camera Depth Estimation](https://mlsanity.com/paper/2304.03369/) (2023)
- [DualRefine: Self-Supervised Depth and Pose Estimation Through Iterative Epipolar Sampling and Refinement Toward Equilibrium](https://mlsanity.com/paper/2304.03560/) (2023)
- [CVRecon: Rethinking 3D Geometric Feature Learning For Neural Reconstruction](https://mlsanity.com/paper/2304.14633/) (2023)
- [Event-based Monocular Dense Depth Estimation with Recurrent Transformers](https://mlsanity.com/paper/2212.02791/) (2022)
- [Reconstructing 4D Spatial Intelligence: A Survey](https://mlsanity.com/paper/2507.21045/) (2025)
- [From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Prior](https://mlsanity.com/paper/2312.10118/) (2023)
- [Crafting Monocular Cues and Velocity Guidance for Self-Supervised Multi-Frame Depth Learning](https://mlsanity.com/paper/2208.09170/) (2022)
- [Zero-Shot Novel View and Depth Synthesis with Multi-View Geometric Diffusion](https://mlsanity.com/paper/2501.18804/) (2025)
- [Two-View Geometry Scoring Without Correspondences](https://mlsanity.com/paper/2306.01596/) (2023)
- [M2Depth: Self-supervised Two-Frame Multi-camera Metric Depth Estimation](https://mlsanity.com/paper/2405.02004/) (2024)
- [Mono-ViFI: A Unified Learning Framework for Self-supervised Single- and Multi-frame Monocular Depth Estimation](https://mlsanity.com/paper/2407.14126/) (2024)
- [Mind The Edge: Refining Depth Edges in Sparsely-Supervised Monocular Depth Estimation](https://mlsanity.com/paper/2212.05315/) (2022)
- [DS-Depth: Dynamic and Static Depth Estimation via a Fusion Cost Volume](https://mlsanity.com/paper/2308.07225/) (2023)
- [Two-in-One Depth: Bridging the Gap Between Monocular and Binocular Self-supervised Depth Estimation](https://mlsanity.com/paper/2309.00933/) (2023)
- [Adjacent-view Transformers for Supervised Surround-view Depth Estimation](https://mlsanity.com/paper/2303.07759/) (2023)
- [ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusion](https://mlsanity.com/paper/2407.09303/) (2024)
- [GRIN: Zero-Shot Metric Depth with Pixel-Level Diffusion](https://mlsanity.com/paper/2409.09896/) (2024)
- [Hybrid-grained Feature Aggregation with Coarse-to-fine Language Guidance for Self-supervised Monocular Depth Estimation](https://mlsanity.com/paper/2510.09320/) (2025)

## Abstract (arXiv)

Multi-frame depth estimation improves over single-frame approaches by also leveraging geometric relationships between images via feature matching, in addition to learning appearance-based features. In this paper we revisit feature matching for self-supervised monocular depth estimation, and propose a novel transformer architecture for cost volume generation. We use depth-discretized epipolar sampling to select matching candidates, and refine predictions through a series of self- and cross-attention layers. These layers sharpen the matching probability between pixel features, improving over standard similarity metrics prone to ambiguities and local minima. The refined cost volume is decoded into depth estimates, and the whole pipeline is trained end-to-end from videos using only a photometric objective. Experiments on the KITTI and DDAD datasets show that our DepthFormer architecture establishes a new state of the art in self-supervised monocular depth estimation, and is even competitive with highly specialized supervised single-frame architectures. We also show that our learned cross-attention network yields representations transferable across datasets, increasing the effectiveness of pre-training strategies. Project page: https://sites.google.com/tri.global/depthformer