mlsanityData to 7 Oct 2026

MobileStereoNet: Towards Lightweight Deep Networks for Stereo Matching

arXiv 2108.09770 ↗First posted 22 Aug 2021cs.CVcited by 82

Not on the map. Why?

Abstract (arXiv)

Recent methods in stereo matching have continuously improved the accuracy using deep models. This gain, however, is attained with a high increase in computation cost, such that the network may not fit even on a moderate GPU. This issue raises problems when the model needs to be deployed on resource-limited devices. For this, we propose two light models for stereo vision with reduced complexity and without sacrificing accuracy. Depending on the dimension of cost volume, we design a 2D and a 3D model with encoder-decoders built from 2D and 3D convolutions, respectively. To this end, we leverage 2D MobileNet blocks and extend them to 3D for stereo vision application. Besides, a new cost volume is proposed to boost the accuracy of the 2D model, making it performing close to 3D networks. Experiments show that the proposed 2D/3D networks effectively reduce the computational expense (27%/95% and 72%/38% fewer parameters/operations in 2D and 3D models, respectively) while upholding the accuracy. Our code is available at https://github.com/cogsys-tuebingen/mobilestereonet.

Key facts. Cited by 82 papers in the map (data through 7 Oct 2026). Builds on 22 of its 32 references that are in the map. First posted on arXiv 22 Aug 2021.

Builds on

22 of 32 references.

References · 22

  1. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  2. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  3. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,054
  4. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  5. 2020X3D: Expanding Architectures for Efficient Video Recognitioncited by 635
  6. 2019Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimationcited by 490
  7. 2018Pyramid Stereo Matching Networkcited by 486
  8. 2017Learning Spatio-Temporal Representation with Pseudo-3D Residual Networkscited by 386
  9. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 359
  10. 2020Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloudcited by 335
  11. 2019GA-Net: Guided Aggregation Net for End-to-end Stereo Matchingcited by 242
  12. 2019Group-wise Correlation Stereo Networkcited by 235
10 more
  1. 2015Stereo Matching by Training a Convolutional Neural Network to Compare Image Patchescited by 151
  2. 2019DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatchcited by 114
  3. 2018Occlusions, Motion and Depth Boundaries with a Generic Network for Disparity, Optical Flow or Scene Flow Estimationcited by 69
  4. 2017Cascade Residual Learning: A Two-stage Convolutional Neural Network for Stereo Matchingcited by 63
  5. 2019Looking at the right stuff: Guided semantic-gaze for autonomous drivingcited by 32
  6. 2019AutoDispNet: Improving Disparity Estimation With AutoMLcited by 29
  7. 2019Fast Deep Stereo with 2D Convolutional Processing of Cost Signaturescited by 12
  8. 20183D Depthwise Convolution: Reducing Model Parameters in 3D Vision Taskscited by 9
  9. 2018CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimationcited by 8
  10. 2017Look Wider to Match Image Patches with Convolutional Neural Networkscited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 6; 2023: 7; 2024: 24; 2025: 29; 2026: 14.

82 papers.

Sort:

By area:

  1. 2023VoxFormer: Sparse Voxel Transformer for Camera-based 3D Semantic Scene Completioncited by 306
  2. 2023Symphonize 3D Semantic Scene Completion with Contextual Instance Queriescited by 88
  3. 2023SSCBench: A Large-Scale 3D Semantic Scene Completion Benchmark for Autonomous Drivingcited by 72
  4. 2024Context and Geometry Aware Voxel Transformer for Semantic Scene Completioncited by 50
  5. 2023Camera-based 3D Semantic Scene Completion with Sparse Guidance Networkcited by 40
  6. 2023OpenStereo: A Comprehensive Benchmark for Stereo Matching and Strong Baselinecited by 28
  7. 2024LightStereo: Channel Boost Is All You Need for Efficient 2D Cost Aggregationcited by 28
  8. 2024A Survey on Deep Stereo Matching in the Twentiescited by 23
  9. 2025Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matchingcited by 23
  10. 2024Event-based Stereo Depth Estimation: A Surveycited by 19
  11. 2021The Implicit Values of A Good Hand Shake: Handheld Multi-Frame Neural Depth Refinementcited by 13
  12. 2024HybridOcc: NeRF Enhanced Transformer-based Multi-Camera 3D Occupancy Predictioncited by 12
38 more
  1. 2025BANet: Bilateral Aggregation Network for Mobile Stereo Matchingcited by 12
  2. 2025VoxDet: Rethinking 3D Semantic Occupancy Prediction as Dense Object Detectioncited by 12
  3. 2025CREStE: Scalable Mapless Navigation with Internet Scale Priors and Counterfactual Guidancecited by 9
  4. 2022ORA3D: Overlap Region Aware Multi-view 3D Object Detectioncited by 8
  5. 2025L2COcc: Lightweight Camera-Centric Semantic Scene Completion via Distillation of LiDAR Modelcited by 8
  6. 2022A Practical Stereo Depth System for Smart Glassescited by 7
  7. 2024Efficient Depth-Guided Urban View Synthesiscited by 7
  8. 2024FIReStereo: Forest InfraRed Stereo Dataset for UAS Depth Perception in Visually Degraded Environmentscited by 7
  9. 2025VisHall3D: Monocular Semantic Scene Completion from Reconstructing the Visible Regions to Hallucinating the Invisible Regionscited by 7
  10. 2022StereoVoxelNet: Real-Time Obstacle Detection Based on Occupancy Voxels from a Stereo Camera Using Deep Neural Networkscited by 6
  11. 2024LightOcc: Lightweight Spatial Embedding for Efficient Vision-based 3D Occupancy Predictioncited by 6
  12. 2025Disentangling Instance and Scene Contexts for 3D Semantic Scene Completioncited by 6
  13. 2024Extending 6D Object Pose Estimators for Stereo Visioncited by 5
  14. 2025OccuFly: A 3D Vision Benchmark for Semantic Scene Completion from the Aerial Perspectivecited by 5
  15. 2022Active-Passive SimStereo -- Benchmarking the Cross-Generalization Capabilities of Deep Learning-based Stereo Methodscited by 4
  16. 2023StereoFlowGAN: Co-training for Stereo and Flow with Unsupervised Domain Adaptationcited by 4
  17. 2024Three Cars Approaching within 100m! Enhancing Distant Geometry by Tri-Axis Voxel Scanning for Camera-based Semantic Scene Completioncited by 4
  18. 2024Drone Stereo Vision for Radiata Pine Branch Detection and Distance Measurement: Integrating SGBM and Segmentation Modelscited by 3
  19. 2024ET-Former: Efficient Triplane Deformable Attention for 3D Semantic Scene Completion From Monocular Cameracited by 3
  20. 2025SGFormer: Satellite-Ground Fusion for 3D Semantic Scene Completioncited by 3
  21. 2025IPFormer: Visual 3D Panoptic Scene Completion with Context-Adaptive Instance Proposalscited by 3
  22. 2025Lite Any Stereo: Efficient Zero-Shot Stereo Matchingcited by 3
  23. 2024Distill-then-prune: An Efficient Compression Framework for Real-time Stereo Matching Network on Edge Devicescited by 2
  24. 2025Out-of-Distribution Semantic Occupancy Predictioncited by 2
  25. 2025StereoCarla: A High-Fidelity Driving Dataset for Generalizable Stereocited by 2
  26. 2021TriStereoNet: A Trinocular Framework for Multi-baseline Disparity Estimationcited by 1
  27. 2022Bayesian Learning for Disparity Map Refinement for Semi-Dense Active Stereo Visioncited by 1
  28. 2023Edge-aware Consistent Stereo Video Depth Estimationcited by 1
  29. 2024VHS: High-Resolution Iterative Stereo Matching with Visual Hull Priorscited by 1
  30. 2024VPOcc: Exploiting Vanishing Point for 3D Semantic Occupancy Predictioncited by 1
  31. 2024Drone Stereo Vision for Radiata Pine Branch Detection and Distance Measurement: Utilizing Deep Learning and YOLO Integrationcited by 1
  32. 2024The Sampling-Gaussian for stereo matchingcited by 1
  33. 2024Efficient Depth Estimation for Unstable Stereo Camera Systems on AR Glassescited by 1
  34. 2025ThermoStereoRT: Thermal Stereo Matching in Real Time via Knowledge Distillation and Attention-based Refinementcited by 1
  35. 2025CurriFlow: Curriculum-Guided Depth Fusion with Optical Flow-Based Temporal Alignment for 3D Semantic Scene Completioncited by 1
  36. 2025Complementary Information Guided Occupancy Prediction via Multi-Level Representation Fusioncited by 1
  37. 2025MAFNet:Multi-frequency Adaptive Fusion Network for Real-time Stereo Matchingcited by 1
  38. 2026Humanoid-OmniOcc: Stereo-Based Full-View Occupancy Dataset for Embodied AIcited by 1

50 most cited of 82 All 82 (JSON)