mlsanityData to 7 Oct 2026

DEVIANT: Depth EquiVarIAnt NeTwork for Monocular 3D Object Detection

arXiv 2207.10758 ↗First posted 21 Jul 2022cs.CV cs.LGcited by 48

Not on the map. Why?

Abstract (arXiv)

Modern neural networks use building blocks such as convolutions that are equivariant to arbitrary 2D translations. However, these vanilla blocks are not equivariant to arbitrary 3D translations in the projective manifold. Even then, all monocular 3D detectors use vanilla blocks to obtain the 3D coordinates, a task for which the vanilla blocks are not designed for. This paper takes the first step towards convolutions equivariant to arbitrary 3D translations in the projective manifold. Since the depth is the hardest to estimate for monocular detection, this paper proposes Depth EquiVarIAnt NeTwork (DEVIANT) built with existing scale equivariant steerable blocks. As a result, DEVIANT is equivariant to the depth translations in the projective manifold whereas vanilla networks are not. The additional depth equivariance forces the DEVIANT to learn consistent depth estimates, and therefore, DEVIANT achieves state-of-the-art monocular 3D detection results on KITTI and Waymo datasets in the image-only category and performs competitively to methods using extra information. Moreover, DEVIANT works better than vanilla networks in cross-dataset evaluation. Code and models at https://github.com/abhi1kumar/DEVIANT

Key facts. Cited by 48 papers in the map (data through 7 Oct 2026). Builds on 81 of its 104 references that are in the map. First posted on arXiv 21 Jul 2022.

Builds on

81 of 104 references.

References · 81

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,617
  5. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  6. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  7. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,436
  8. 2020YOLOv4: Optimal Speed and Accuracy of Object Detectioncited by 1,624
  9. 2019Objects as Pointscited by 1,172
  10. 2016Group Equivariant Convolutional Networkscited by 1,154
  11. 2021Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gaugescited by 1,131
  12. 2015Multi-Scale Context Aggregation by Dilated Convolutionscited by 1,128
38 more
  1. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 939
  2. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 735
  3. 2021DETR3D: 3D Object Detection from Multi-view Images via 3D-to-2D Queriescited by 655
  4. 2018Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Drivingcited by 455
  5. 2018Spherical CNNscited by 452
  6. 2017Deep Layer Aggregationcited by 436
  7. 2018Hyperbolic Neural Networkscited by 393
  8. 2019From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimationcited by 379
  9. 2021Categorical Depth Distribution Network for Monocular 3D Object Detectioncited by 326
  10. 2016Harmonic Networks: Deep Translation and Rotation Equivariancecited by 293
  11. 2019M3D-RPN: Monocular 3D Region Proposal Network for Object Detectioncited by 251
  12. 2021Is Pseudo-Lidar needed for Monocular 3D Object detection?cited by 248
  13. 2019Disentangling Monocular 3D Object Detectioncited by 215
  14. 2017Learning Steerable Filters for Rotation Equivariant CNNscited by 193
  15. 2021Objects are Different: Flexible Monocular 3D Object Detectioncited by 184
  16. 2019Learning Depth-Guided Convolutions for Monocular 3D Object Detectioncited by 169
  17. 2021Geometry Uncertainty Projection Network for Monocular 3D Object Detectioncited by 155
  18. 2020MonoPair: Monocular 3D Object Detection Using Pairwise Spatial Relationshipscited by 153
  19. 2020RTM3D: Real-time Monocular 3D Detection from Object Keypoints for Autonomous Drivingcited by 150
  20. 2020Train in Germany, Test in The USA: Making 3D Object Detectors Generalizecited by 150
  21. 2021Delving into Localization Errors for Monocular 3D Object Detectioncited by 138
  22. 2019Accurate Monocular Object Detection via Color-Embedded 3D Reconstruction for Autonomous Drivingcited by 137
  23. 2020Incorporating Symmetry into Deep Dynamics Models for Improved Generalizationcited by 123
  24. 2017Augmented Reality Meets Computer Vision : Efficient Data Generation for Urban Driving Scenescited by 122
  25. 2021MSeg: A Composite Dataset for Multi-domain Semantic Segmentationcited by 122
  26. 2020Kinematic 3D Object Detection in Monocular Videocited by 116
  27. 2019Deep Scale-spaces: Equivariance Over Scalecited by 113
  28. 2019Scale-Equivariant Steerable Networkscited by 110
  29. 2020Rethinking Pseudo-LiDAR Representationcited by 109
  30. 2016Rotation equivariant vector field networkscited by 98
  31. 2017Deep MANTA: A Coarse-to-fine Many-Task Network for joint 2D and 3D vehicle analysis from monocular imagecited by 95
  32. 2016Exploiting Cyclic Symmetry in Convolutional Neural Networkscited by 91
  33. 2020LUVLi Face Alignment: Estimating Landmarks' Location, Uncertainty, and Visibility Likelihoodcited by 87
  34. 2021AutoShape: Real-Time Shape-Aware Monocular 3D Object Detectioncited by 87
  35. 2021Depth-conditioned Dynamic Message Propagation for Monocular 3D Object Detectioncited by 84
  36. 2021Geometry-based Distance Decomposition for Monocular 3D Object Detectioncited by 83
  37. 2017Polar Transformer Networkscited by 81
  38. 2020Meta-Learning Symmetries by Reparameterizationcited by 75

50 most cited of 81 All 81 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 15; 2024: 14; 2025: 11; 2026: 6.

48 papers.

Sort:

By area:

  1. 2024TUMTraf V2X Cooperative Perception Datasetcited by 66
  2. 2024MonoCD: Monocular 3D Object Detection with Complementary Depthscited by 38
  3. 2023Learning Occupancy for Monocular 3D Object Detectioncited by 35
  4. 2023Tame a Wild Camera: In-the-Wild Monocular Camera Calibrationcited by 34
  5. 2023MonoNeRD: NeRF-like Representations for Monocular 3D Object Detectioncited by 33
  6. 2024MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priorscited by 21
  7. 2023MonoLSS: Learnable Sample Selection For Monocular 3D Detectioncited by 18
  8. 2024MonoMAE: Enhancing Monocular 3D Detection through Depth-Aware Masked Autoencoderscited by 15
  9. 2023NeurOCS: Neural NOCS Supervision for Monocular 3D Object Localizationcited by 13
  10. 2022CrossDTR: Cross-view and Depth-guided Transformers for 3D Object Detectioncited by 10
  11. 2024MonoTAKD: Teaching Assistant Knowledge Distillation for Monocular 3D Object Detectioncited by 10
  12. 2023SSD-MonoDETR: Supervised Scale-aware Deformable Transformer for Monocular 3D Object Detectioncited by 9
36 more
  1. 2024Depth-discriminative Metric Learning for Monocular 3D Object Detectioncited by 8
  2. 2025Unleashing the Power of Chain-of-Prediction for Monocular 3D Object Detectioncited by 8
  3. 20233D Video Object Detection with Learnable Object-Centric Global Optimizationcited by 7
  4. 2023Monocular 3D Object Detection with Bounding Box Denoising in 3D by Perceivercited by 7
  5. 2024Multi-View Attentive Contextualization for Multi-View 3D Object Detectioncited by 7
  6. 2024Multi-task Learning for Real-time Autonomous Driving Leveraging Task-adaptive Attention Generatorcited by 6
  7. 2023GUPNet++: Geometry Uncertainty Propagation Network for Monocular 3D Object Detectioncited by 5
  8. 2024TiGDistill-BEV: Multi-view BEV 3D Object Detection via Target Inner-Geometry Learning Distillationcited by 4
  9. 2023Improving Equivariance in State-of-the-Art Supervised Depth and Normal Predictorscited by 3
  10. 2024Current Symmetry Group Equivariant Convolution Frameworks for Representation Learningcited by 2
  11. 2025MonoSOWA: Scalable monocular 3D Object detector Without human Annotationscited by 2
  12. 2025MonoCT: Overcoming Monocular 3D Detection Domain Shift with Consistent Teacher Modelscited by 2
  13. 2025StereoDETR: Stereo-based Transformer for 3D Object Detectioncited by 2
  14. 2025IDEAL-M3D: Instance Diversity-Enriched Active Learning for Monocular 3D Detectioncited by 2
  15. 2026Towards Intrinsic-Aware Monocular 3D Object Detectioncited by 2
  16. 2023Augment and Criticize: Exploring Informative Samples for Semi-Supervised Monocular 3D Object Detectioncited by 1
  17. 2023Parcel3D: Shape Reconstruction from Single RGB Images for Applications in Transportation Logisticscited by 1
  18. 2023Towards Fair and Comprehensive Comparisons for Image-Based 3D Object Detectioncited by 1
  19. 2024LAM3D: Leveraging Attention for Monocular 3D Object Detectioncited by 1
  20. 2025AuxDepthNet: Real-Time Monocular 3D Object Detection with Depth-Sensitive Featurescited by 1
  21. 2024AI-Powered CPS-Enabled Vulnerable-User-Aware Urban Transportation Digital Twin: Methods and Applicationscited by 1
  22. 2025Hybrid Rendering for Multimodal Autonomous Driving: Merging Neural and Physics-Based Simulationcited by 1
  23. 2025All You Need for Object Detection: From Pixels, Points, and Prompts to Next-Gen Fusion and Multimodal LLMs/VLMs in Autonomous Vehiclescited by 1
  24. 2025LeAD-M3D: Leveraging Asymmetric Distillation for Real-Time Monocular 3D Detectioncited by 1
  25. 2026DepthAgent: Towards Better Universal Depth Estimation via Sample-wise Expert Selectioncited by 1
  26. 2022Recursive Cross-View: Use Only 2D Detectors to Achieve 3D Object Detection without 3D Annotationscited by 0
  27. 2023OriCon3D: Effective 3D Object Detection using Orientation and Confidencecited by 0
  28. 2023S3-MonoDETR: Supervised Shape&Scale-perceptive Deformable Transformer for Monocular 3D Object Detectioncited by 0
  29. 2024Efficient Feature Aggregation and Scale-Aware Regression for Monocular 3D Object Detectioncited by 0
  30. 2024Revisiting Monocular 3D Object Detection with Depth Thickness Fieldcited by 0
  31. 2025S-LAM3D: Segmentation-Guided Monocular 3D Object Detection via Feature Space Fusioncited by 0
  32. 2025MAESTRO: Task-Relevant Optimization via Adaptive Feature Enhancement and Suppression for Multi-task 3D Perceptioncited by 0
  33. 2026Mono3DV: Monocular 3D Object Detection with 3D-Aware Bipartite Matching and Variational Query DeNoisingcited by 0
  34. 2026Selective Transfer Learning of Cross-Modality Distillation for Monocular 3D Object Detectioncited by 0
  35. 2026MonoPRIO: Adaptive Prior Conditioning for Unified Monocular 3D Object Detectioncited by 0
  36. 2026RefineAny3D: Depth Refinement as Semantic Alignment for Monocular 3D Detectioncited by 0