The DOI matches its Crossref record; 66 of the 82 references it cites were found in a registry or an index.
Categorical Depth Distribution Network for Monocular 3D Object Detection
Open accessAbstract
Monocular 3D object detection is a key problem for autonomous vehicles, as it provides a solution with simple configuration compared to typical multi-sensor systems. The main challenge in monocular 3D detection lies in accurately predicting object depth, which must be inferred from object and scene cues due to the lack of direct range measurement. Many methods attempt to directly estimate depth to assist in 3D detection, but show limited performance as a result of depth inaccuracy. Our proposed solution, Categorical Depth Distribution Network (CaDDN), uses a predicted categorical depth distribution for each pixel to project rich contextual feature information to the appropriate depth interval in 3D space. We then use the computationally efficient bird’s-eye-view projection and single-stage detector to produce the final output detections. We design CaDDN as a fully differentiable end-to-end approach for joint depth estimation and object detection. We validate our approach on the KITTI 3D object detection benchmark, where we rank 1stamong published monocular methods. We also provide the first monocular 3D detection results on the newly released Waymo Open Dataset. We provide a code release for CaDDN which is made available.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“A popular BEV framework that generates BEV features is based on depth information [46, 32, 34], but this paradigm is sensitive to the accuracy of depth values or the depth distributions.”
From BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers (2022) · cites this paper as [34]
In the citing paper
“BEV-based approaches have not shown significant advantages over other paradigm in 3D object detections [47, 31, 34].”
From BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers (2022) · cites this paper as [34]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
Fully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
PyTorch: An Imperative Style, High-Performance Deep Learning Library
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al. (2019) · arXiv · doi:10.48550/arxiv.1912.01703 · Cited by 31,473
Focal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.324 · Cited by 30,581
Chen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2699184 · Cited by 22,392
Chen, L.-C., Papandreou, G., Kokkinos, I., Murphy, K., & Yuille, A. L. (2016) · arXiv · doi:10.48550/arxiv.1606.00915 · Cited by 22,392
Are we ready for autonomous driving? The KITTI vision benchmark suite
Geiger, A., Lenz, P., & Urtasun, R. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6248074 · Cited by 14,897
Focal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2858826 · Cited by 9,556
Flagged first, then by citations
16 of the 82 references couldn’t be matched to a known paper.
Research works printed but not linked: 4
- mani, 2020, Monolayout: Amodal scene layout from a single image, WACV
- ma, 2020, Rethinking pseudolidar representation, ECC
- shi, 2020, Distance-normalized unified representation for monocular 3d object detection, ECC
- roddick, 2018, Orthographic feature transform for monocular 3D object detection, BMV
Cited by
Sorted by citations
Items 401–500 of 597
Semantic Shape and Trajectory Reconstruction for Monocular Cooperative 3D Object Detection
Cserni, M., & Rövid, A. (2024) · IEEE Access · doi:10.1109/access.2024.3484672 · Cited by 2
“Deep learning based methods such as [1], [2], [3], [4], [5], [6], [7], [22], and [23] do not estimate vehicle shape, and do not perform well without context-specific retraining.”
Yang, K., Wang, Y., Han, Y., & Jia, Q. (2024) · IEEE International Conference on Automation Science and Engineering (CASE) · doi:10.1109/case59546.2024.10711318 · Cited by 2
“The bottom-up strategy [13]–[15] involves projecting the 2D image into 3D space via depth estimation and converting it into a BEV mesh.”
BackgroundConstructing 3D Object Detectors Based on Deformable Convolutional Guided Depths
Zheng, X., Yang, G., Yang, L., & Lu, C. (2024) · IEEE Access · doi:10.1109/access.2024.3488748 · Cited by 2
“In addition, CaDDN [29] constructs a bird’s eye view (BEV) representation by learning the classification depth distribution of each pixel and recovering the enclosing box from the BEV projection.”
BackgroundRecurrent Cross-Modality Fusion for Time-of-Flight Depth Denoising
Dong, G., Zhang, Y., Sun, X., & Xiong, Z. (2024) · IEEE Transactions on Computational Imaging · doi:10.1109/tci.2024.3496312 · Cited by 2
“Despite the success of ToF cameras in the market, they are still plagued by noise, which can degrade their performance in high-level tasks [5], [6], [7], [8], [9], [10], [11].”
FusionReg: LiDAR-Camera Fusion Regression Enhancement for 3D Object Detection
Rongchun, X. I. A. O., Yuansheng, L. I. U., Zhang, J., HUANG, Y., & Han, X. (2024) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2024edp7158 · Cited by 2
“Moreover CaDDN[22] employed interpolation to supervise the training of depth estimation using discrete depth maps from LiDAR data.”
Improved Pseudo-Stereo for Monocular 3D Object Detection in Autonomous Driving
Lin, T., Zhang, Y., & Zhu, Y. (2024) · International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS) · doi:10.1109/iciibms62405.2024.10792823 · Cited by 2
“Our method outperforms MonoFlex[13], CaDDN[14], MonoDTR[15], and MonoRCNN++[16] at IOU ≥ 0.7 for the evaluation metrics AP_3D for 3D object detection and AP_BEV for bird's-eye view detection across three difficulty levels: easy, medium, and hard.”
Influential3D Object Detection via Stereo Pyramid Transformers with Rich Semantic Feature Fusion
Gu, R., Yang, C., Lu, Y., Liu, P., Wu, F. Q., & Chen, G. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802137 · Cited by 2
“2D feature-based approaches, like LSS [12], estimate depth distribution and generate 3D frustum features, increasingly popular in recent studies [13]–[16].”
OC-BEVDepth: An Object Contour-Perceiving 3D BEV Object Detection Framework
Chen, X., Pei, X., Liu, Z., & Zhang, Y. (2024) · CAA International Conference on Vehicular Control and Intelligence (CVCI) · doi:10.1109/cvci63518.2024.10830140 · Cited by 2
A 3D Object Detection Method Based on Cross-Attention Multimodal Feature Fusion
Tian, J., Zhou, Y., Peng, M., & Chen, X. (2024) · CAA International Conference on Vehicular Control and Intelligence (CVCI) · doi:10.1109/cvci63518.2024.10830167 · Cited by 2
“The compared algorithms include: depth based monocular pure visual object detection method CaDDN [5], VoxelNet [14] based on point cloud voxelization expression, Autoalign [15] based on early fusion of image features and voxel features, and PI-RCNN [16] directly fused in point clouds.”
Obstalaneyolo: Real-Time Lane and Obstacle Detection for Autonomous Vehicles
Manzoor, M., Li, J., Zhou, L., Zeng, A., Abbas, M. H., Tabraiz, S., & Hassan, I. (2024) · International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) · doi:10.1109/iccwamtip64812.2024.10873687 · Cited by 3
A scalable monocular 3D detector with Superpixel Feature Pyramid Network
Ma, D., Zhao, F., Li, Y., Qu, X., Jiang, X., Wu, H., Chen, X., & Liu, M. (2025) · Applied Soft Computing · doi:10.1016/j.asoc.2025.113389 · Cited by 2
Fusing depth embeddings for monocular 3D object detection
Ji, C., Zhao, D., Li, W., & Liu, G. (2025) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2025.104555 · Cited by 2
Unified Human Localization and Trajectory Prediction with Monocular Vision
Luan, P., Gao, Y., Demonsant, C., & Alahi, A. (2025) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra55743.2025.11128645 · Cited by 2
“While LiDAR data can enhance performance in 3D object detection [14], [15], its high cost limits its accessibility.”
Data Scaling Laws for End-to-End Autonomous Driving
Naumann, A., Gu, X., Dimlioglu, T., Bojarski, M., Değirmenci, A., Popov, A., Bisla, D., Pavone, M., Müller, U., & Ivanovic, B. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00242 · Cited by 2
“Recent advancements in RGB camera-based approaches have achieved competitive results across 3D object detection [32, 58, 69, 78], HD map construction [44–47, Building on this, the outputs of perception tasks serve as critical inputs for downstream modules, such as trajectory prediction and planning.”
Aliasing fringe pattern denoising based on deep learning
王, 小., 于, 瀛., & 胡, 健. (2021) · doi:10.1117/12.2606566 · Cited by 2
“Deep learning, as a subfield of machine learning, is widely used in multiple fields such as speech recognition [4] and target detection [5] by learning the properties of data through multilayer networks.”
BackgroundFengsui, 王. W., Lei, 熊. X., & Yaping, 钱. Q. (2023) · Laser & Optoelectronics Progress · doi:10.3788/lop222627 · Cited by 2
Sun, H., Fan, Z., Song, Z., Wang, Z., Wu, K., & Lu, J. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3413179 · Cited by 2
“In our work, we apply MonoSIM to the famous M3D-RPN [10], CaDDN [20], and GUPNet [47] detectors.”
BackgroundMethodInfluentialMonocular 3D object detection with thermodynamic loss and decoupled instance depth
Liu, G., Xie, X., & Yu, Q. (2024) · Connection Science · doi:10.1080/09540091.2024.2316022 · Cited by 1
“CaDDN (Reading et al., 2021) obtains more accuratedepthestimationbypredictingthecategoryofeachobjectandthedepthdistribu-tion within each category, and uses the gridded depth prediction to generate high-quality BEV images for 3D detection.”
DVT: Decoupled Dual-Branch View Transformation for Monocular Bird’s Eye View Semantic Segmentation
Du, J., Pan, X., Shen, M., Su, S., Yang, J., Liu, C., & Chen, Q. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802126 · Cited by 1
“Subsequent works [12], [13] follow this LSS paradigm as well.”
Object-level Data Augmentation for Visual 3D Object Detection in Autonomous Driving
Cheng, H., Xu, J., Peng, L., Yang, Z., He, X., & Wu, B. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10887542 · Cited by 1
Depth-free view synthesis from diffusion models for monocular 3D detector in autonomous driving
Shi, Y., Zhou, S., Wang, W., & Lu, X. (2025) · Multimedia Systems · doi:10.1007/s00530-025-01819-0 · Cited by 1
Pedestrian dynamics modeling and social force analysis based on object detection
Yang, F., Liu, R., & Zhu, D. (2025) · Frontiers in Physics · doi:10.3389/fphy.2025.1579280 · Cited by 1
BEVRoad: A Cross-Modal and Temporary-Recurrent 3D Object Detector for Infrastructure Perception
Li, X., Zhang, J., Gu, J., Lu, X., & Zhang, L. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-6594-5_20 · Cited by 1
Advancing 3D Object Detection With Depth-Aware Spatial Knowledge Distillation
Wu, Z., Song, F., Gan, Y., Wu, Y., Xu, T., Wang, X., Tang, R., & Pu, J. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3594805 · Cited by 1
“CaDDN [1] proposed categorical depth distribution learning.”
InfluentialRevisiting Cross-Domain Problem for LiDAR-Based 3D Object Detection
Zhang, R., Lee, J., Cai, X., & Prügel‐Bennett, A. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-7036-9_6 · Cited by 1
MonoTrans: Long-Range Contextual Dependency Transformer for Monocular 3-D Object Detection
Li, Y., Liu, X., Hua, Z., & Sun, Z.-W. (2025) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2025.3593332 · Cited by 1
CatFree3D: Category-Agnostic 3D Object Detection with Diffusion
Bian, W., Wang, Z.-R., & Vedaldi, A. (2025) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv66043.2025.00015 · Cited by 2
“Depending on the dataset domain, certain models are tailored for outdoor self-driving scenes [11, 23, 35, 37, 39, 39, 44, 45, 61], while others are specifically designed for indoor environments [24, 30, 40, 59].”
Research on Indoor 3D Semantic Mapping Based on ORB-SLAM2 and Multi-Object Tracking
Wang, W., Wu, R., Dong, Y., & Jiang, H. (2025) · Applied Sciences · doi:10.3390/app152010881 · Cited by 1
Local Vectorized High Definition Map Construction for Autonomous Driving: A Comprehensive Review
Zhang, Y., Qian, Y., Meng, C., Zhang, R., Yi, H., Wang, C., & Yang, M. Z. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3623833 · Cited by 1
“CaDDN [73] adopts a similar approach but uses real depth values to supervise the network, improving depth estimation.”
Iter3DDet: Depth-Guided Iterative Fusion and Refinement for Monocular 3D Object Detection
Feng, C., Zhang, C., Chen, Z., Hu, W., Lü, K., & Ge, L. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3629741 · Cited by 1
“This approach was extended by CaDDN [26], which pioneered di ff erentiable depth discretization via categorical depth distributions to generate optimized bird’s-eye view features for precise localization.”
PRTF: Polar Space Represented Multi-View 3D Object Detection With Temporal Fusion Enhancement
Tang, J., Hou, Y., Liu, J., & Bo, Y. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3633448 · Cited by 2
“V: first is the simple Uniform Division (UD), second is the Linear Increasing Discretization (LID) proposed in CaDDN [53], LID can be defined by: where [ d min , d max ] is depth range, D is the number of depth divisions, and d i is the index of the depth location.”
A Lightweight Multimodal Fusion Method for Object Detection Based on Bird’s Eye View
Chen, H., Chang, S., Ye, W., Gu, D., Xu, Q., & Ji, M. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3639093 · Cited by 1
“Both LSS [18] and CaDDN [19] adopt a typical 2-”
Test-Time Correction: An Online 3D Detection System via Visual Prompting
Zhang, H., Yang, Z., Sun, Y., Chen, L., Xia, F., Güney, F., & Li, H. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3642076 · Cited by 1
“Existing 3D object detectors [5], [6], [7], [8], [9] typically follow an offline training and deployment pipeline.”
MonoGSDet: Monocular 3D Object Detection With Gaussian Splatting in Autonomous Driving
Li, T., & Liu, M. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3621441 · Cited by 1
“For instance, CaDDN [11] converts the frustum into voxel space and then collapses it into a bird’s-eye view (BEV) representation.”
PARTNER: Level Up the Polar Representation for 3D Object Detection
Nie, M., Wang, C., Xu, H., & Zhang, L. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-026-02735-0 · Cited by 1
Uncertainty-Aware Projection Network for Multi-View 3D Object Detection
Jiang, H., Yang, W., Zhang, T., & Zhang, Y. (2026) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2026.3675364 · Cited by 2
“CaDDN [29] jointly optimize accurate depth estimation as well as 3D object detection, leading an increasing performance.”
A stable diffusion-based method for low-light image enhancement in autonomous driving
Zhang, Y., Hui, J., & Du, Y. (2026) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-026-03053-0 · Cited by 1
MonoGATR: Monocular 3D object detection with a geometry-aware transformer
Xie, C., Song, C., Chen, H., & Li, Y. (2026) · Displays · doi:10.1016/j.displa.2026.103480 · Cited by 1
He, Y., Zhao, L., Deng, T., Fang, Z., & Chen, W. (2026) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2026.3687354 · Cited by 1
“Following LSS [8], camera-based methods gained popularity, with both monocular [30] and multiview [31], [32] approaches improving detection accuracy.”
CHARM3R: Towards Unseen Camera Height Robust Monocular 3D Detector
Kumar, A., Guo, Y., Zhang, Z., Huang, X., Ren, L., & Liu, X. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00821 · Cited by 1
“A few use NMS [40, 55], corrected extrinsics [120], CAD models [9, 43, 58] or LiDAR [82] in training.”
IDEAL-M3D: Instance Diversity-Enriched Active Learning for Monocular 3D Detection
Meier, J., Günther, F., Marin, R., Dhaouadi, O., Kaiser, J., & Cremers, D. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00026 · Cited by 3
“Following the monocular setting [45], we use front-camera images with 52,386 training and 39,848 validation images.”
InfluentialModeling and Learning Multiple Hypotheses for Monocular 3D Object Detection
Park, H., Xiong, P., Yu, P., & Tang, W. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00680 · Cited by 1
“The k -th hypothesis is represented as where ( x k , y k ) , d k , s k , θ k , and c k are the k -th hypothesis’s x and y-coordinates in 3D space, depth, 3D size, orientation, and confidence, respectively.”
Cipelli, L., D'Addeo, F., Ghelfi, E., Ceresini, M., Bertogalli, A., Pirazzoli, F., & Bertozzi, M. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-026-02921-0 · Cited by 1
“For each image pixel [ u , v ], we sample the depth information D using the Linear-Increasing Discretization (LID) (Reading et al., 2021), building the set of coordinates I ∈ R 3 × H / 16 × W / 16 × D .”
MP-Mono: Monocular 3D Detection Using Multiple Priors for Autonomous Driving
Liu, P., Yang, G., Li, P., Wang, Z., Shi, J., Deng, Z., & Qiao, Y. (2021) · International Conference on 3D Vision (3DV) · doi:10.1109/3dv53792.2021.00063 · Cited by 1
“CaDDN [39] constructs frustum features from categorical depth distribution and further transforms them into voxel features.”
BackgroundDetecting Obstacle in 3D Space using Monocular Camera
Haq, M. A., Ruan, S.-J., & Chen, J.-H. (2022) · IEEE Global Conference on Life Sciences and Technologies (LifeTech) · doi:10.1109/lifetech53646.2022.9754879 · Cited by 1
Quantity over Quality: Training an AV Motion Planner with Large Scale Commodity Vision Data
Platinsky, L., Naseer, T., Chen, H., Haines, B., Zhu, H., Grimmett, H., & Pero, L. D. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9982116 · Cited by 1
“methods exist also to process monocular camera data [27], the system we use is built on methods specialised for stereo inputs [28].”
MethodOvercoming the Seesaw in Monocular 3D Object Detection Via Language Knowledge Transferring
Xu, W., & Fu, T. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10095283 · Cited by 1
Li, X., Ye, Z., Wei, S., Chen, Z., Chen, X., Tian, Y., Dang, J., Fu, S., & Zhao, Y. (2023) · Journal of Image and Graphics · doi:10.11834/jig.230036 · Cited by 4
Li, K., Wu, X., Zhang, W., & Yu, W. (2023) · World Electric Vehicle Journal · doi:10.3390/wevj14090239 · Cited by 1
Monocular 3D object detection for distant objects
Li, J., & Han, X. (2024) · Journal of Electronic Imaging · doi:10.1117/1.jei.33.3.033021 · Cited by 1
“14 to 18, have successfully implemented pseudo-LiDAR-based monocular 3D object detection, achieving commendable accuracy.”
BackgroundMethodA streamlined framework for BEV-based 3D object detection with prior masking
Tong, Q., Zhang, J., Yan, C. C., & Zeng, D. (2024) · Image and Vision Computing · doi:10.1016/j.imavis.2024.105229 · Cited by 1
ATV3D: 3D Object Detection from Attention-based Three-view Representation
Han, Y., Chen, Y., Li, H., Zhao, Y., Zhao, Z., & Hu, P. (2024) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn60899.2024.10651302 · Cited by 1
Sun, R., Fu, M., Liang, H. Z., Zhu, C., Dong, Z., & Yang, Y. M. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10802840 · Cited by 1
“Some methods derive depth from pre-training [20]–[22], whereas others embed depth estimation within the network to facilitate simultaneous training [15], [23], [24].”
BEVPose: Unveiling Scene Semantics through Pose-Guided Multi-Modal BEV Alignment
Hosseinzadeh, M., & Reid, I. D. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10801390 · Cited by 1
“Additionally, works like [31] and LSS [32] have made progress in camera-based 3D perception, converting camera features to BEV and extending existing models for 3D object detection.”
Ret3D: rethinking object relations for efficient 3D object detection
Wu, Y., ZHANG, D., Liu, Y., ZHANG, L., & Cheng, M. (2025) · Scientia Sinica Informationis · doi:10.1360/ssi-2024-0295 · Cited by 1
Weak Cube R-CNN: Weakly Supervised 3D Detection Using Only 2D Bounding Boxes
Hansen, A. L., Wanzeck, L., & Papadopoulos, D. P. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-95918-9_10 · Cited by 1
UniCuboid: Cuboid-based dense shape supervision for monocular 3D object detection
Zhang, X., Su, Y., Shi, H., Lu, H., Liu, Y., & Wang, L. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130873 · Cited by 1
MDE-Net: Monocular Distance measurement based on detected license plates
Zha, Y., Li, F., Li, Y., Hong, L., & Ling, Q. (2022) · Chinese Control Conference (CCC) · doi:10.23919/ccc55666.2022.9902166 · Cited by 1
LiteDepth: Digging into Fast and Accurate Depth Estimation on Mobile Devices
Li, Z., Chen, Z., Xu, J., Liu, X., & Jiang, J. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-25063-7_31 · Cited by 1
Shape Prior Non-Uniform Sampling Guided Real-time Stereo 3D Object DetectionPreprint
Gao, A., Cao, J., & Pang, Y. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.10013 · Cited by 1
“CaDDN [34] uses a predicted categorical depth distribution for each”
MethodScalable Primitives for Generalized Sensor Fusion in Autonomous VehiclesPreprint
Sidhu, S., Wang, L., Naseer, T., Malhotra, A., Chia, J., Ahuja, A., Rasmussen, E., Huang, Q., & Gao, R. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2112.00219 · Cited by 1
“CaDDN [24] copes with this problem by estimating depth distribution for each pixel before projecting the features to 3D.”
BackgroundLXL: LiDAR Excluded Lean 3D Object Detection With 4D Imaging Radar and Camera FusionPreprint
Xiong, W., Liu, J., Huang, T., Han, Q., Xia, Y., & Zhu, B. (2023) · arXiv (Cornell University) · doi:10.25916/sut.26254550.v1 · Cited by 1
M&M3D: Multi-Dataset Training and EfficientNetwork for Multi-view 3D Object DetectionPreprint
Zhang, H. (2024) · Research Square · doi:10.21203/rs.3.rs-4592321/v1 · Cited by 1
Monocular 3D Object Detection Based on Occlusion Optimization
Fang, L., Huang, Y., Zhao, Q., & Wan, Y. (2022) · International Conference on Intelligent Computing and Signal Processing (ICSP) · doi:10.1109/icsp54964.2022.9778720
Ngo, D. T., Bui, M.-Q. V., Nguyen, D. D., & Pham, H.-A. (2022) · PeerJ Computer Science · doi:10.7717/peerj-cs.1144
Learning Implicit Depth Information for Monocular 3D Object Detection
Weber, M., Polley, R., & Zöllner, J. M. (2022) · International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME) · doi:10.1109/iceccme55909.2022.9988525
“In [21] the categorical BEV depth is estimated to generate BEV features which are then used for 3D object detection.”
MethodLAPTNet: LiDAR-Aided Perspective Transform Network
Diaz-Zapata, M., Erkent, Ö., Laugier, C., Dibangoye, J., & González, D. (2022) · International Conference on Control, Automation, Robotics and Vision (ICARCV) · doi:10.1109/icarcv57592.2022.10004254
“Cameras have been one of the most widely used sensors for tasks such as image segmentation, 2D and 3D object detection [1], [2], [3].”
MethodPainted PIXOR: A Real-Time Multi-Sensor Fusion Object Detection Pipeline for Autonomous Vehicles
Liu, D., Hua, G., & Mo, L. (2022) · Youth Academic Annual Conference of Chinese Association of Automation (YAC) · doi:10.1109/yac57282.2022.10023903
“We can divide these monocular methods into three categories with the consideration of representation types: direct, depth-based, and grid-based methods [14].”
MethodBADE: BEV-Adaptive Distance Estimation Neural Network based on Monocular Cameras
Zha, Y., Li, F., Yin, X., Xu, Y., Cao, L., & Ling, Q. (2022) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc55256.2022.10033725
“Recent years, researchers paid a lot of attention on imaged-based monocular 3D object detection [11, 5, 13], where the location of objects can be obtained from the output of the object detection network and the depth of objects is one dimension of the obtained location.”
BackgroundAn Algorithm on Monocular 3D Object Detection Based on Depth Estimation
Liu, Q., Li, W., Yu, S., Song, Y., Zhou, Q., & Zou, W. (2023) · International Conference on Intelligent Control, Measurement and Signal Processing (ICMSP) · doi:10.1109/icmsp58539.2023.10171096
“CaDDN[3] also transformed the images to the frustum features, followed by the cubic characteristics via the camera, and finally, applied [2] to perform BEV view transformation and related detection.”
Method3D object detection with low resolution depth map
Huang, W., Chen, X.-Z., Chen, Y., & Huang, C.-S. (2023) · IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan) · doi:10.1109/icce-taiwan58799.2023.10226909
“The architecture of CaDDN[5] This research was financially sponsored by the National Science and Technology Council of the Republic of China under the contract No.”
BackgroundMethodInfluentialDASANet: A 3D Object Detector with Density-and-Sparsity Feature Aggregation
Zhang, Q., & Wei, D. (2023) · Remote Sensing · doi:10.3390/rs15184587
DMFF: dual-way multimodal feature fusion for 3D object detection
Dong, X., Di, X., & Wang, W. (2023) · Signal Image and Video Processing · doi:10.1007/s11760-023-02772-z
“CaDDN [26] learns to generate BEV representations from images by predicting a set of depth distributions for each pixel.”
3D Object Detection in Autonomous Driving
Peng, Y., Liu, Y., Yan, X., Li, J., Wang, J., Tai, L., Jin, N., Fan, R., & Liu, M. (2023) · Advances in computer vision and pattern recognition · doi:10.1007/978-981-99-4287-9_5
Collaborative 3D Object Detection
Chen, S., & Hu, Y. (2023) · Advances in computer vision and pattern recognition · doi:10.1007/978-981-99-4287-9_6
Object Detection with Depth Information in Road Scenes
Liu, R., Chen, X., & Tao, B. (2023) · Communications in computer and information science · doi:10.1007/978-981-99-8021-5_15
Attention multilayer feature fusion network for 3D object detection
Zhang, B., Su, C., & Cao, G. (2023) · International Conference on Information Systems Engineering (ICISE) · doi:10.1109/icise60366.2023.00130
“The methods proposed in [10-13] are based on Pseudo-LiDAR.”
MethodSODet: A LiDAR-Based Object Detector in Bird’s-Eye View
Jin, P., & Zhou, Y. (2023) · Communications in computer and information science · doi:10.1007/978-981-99-8148-9_7
Object Recognition and Spatial Relationship Determination in Point Clouds
Song, Q., & Kubota, N. (2023) · International Conference on Machine Learning and Cybernetics · doi:10.1109/icmlc58545.2023.10327997
“Among them, PointNet, PointNet++, DGCNN, RSNet, and other methods have achieved good results[10].”
MethodHistory-Detr: Optimize Query Initialization Strategy by Using Historical Information and Kinematics
Luo, W., Liu, Z. W., Dai, G., & Xu, N. (2023) · ACM International Conference on Multimedia in Asia (MMAsia) · doi:10.1145/3595916.3626373
“They[8, 10, 29]predict adistribution overdepth binsand thenlift imagefeatures to a 3D point cloud.”
Background3D Object Detection Algorithm Based on Multi-Feature Fusion from the Perspective of BEV
Zhou, Y., Peng, C., & Sun, R. (2023) · International Conference on Advances in Artificial Intelligence and Applications (AAIA) · doi:10.1145/3603273.3635235
A 3D object detection strategy based on YOLOv8 combined with CaDDN for vehicle images
Liu, S., Xu, Z., Wang, F., & Qiu, Z. (2023) · International Conference on Image Processing, Computer Vision and Machine Learning (ICICML) · doi:10.1109/icicml60161.2023.10424808
“…in unmanned field, this paper designs a 3D object detection method for monocular vehicle images based on YOLOV8 [4] as well as CaDDN method [5], which classifies and obtains the 2D bounding box of the vehicle through YOLOv8 model, and the 3D bounding box is obtained by depth estimation…”
MethodInfluentialAn Acceleration Inference Implementation of BEVFusion with MQBench on Xavier
Long, Y., Zan, X., Xue, J., & Mei, K. (2023) · China Automation Congress (CAC) · doi:10.1109/cac59555.2023.10451793
“After LSS [1], CaDNN [2] follows a similar path to predict depth.”
BackgroundMethodOladele, D. A., Markus, E. D., & Abu-Mahfouz, A. M. I. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4772696
Multi-Scale Enhanced Depth Knowledge Distillation for Monocular 3D Object Detection with SEFormer
Zhang, H., Li, J., Tang, R. H., Shi, Z., & Bu, A. (2023) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/ithings-greencom-cpscom-smartdata-cybermatics60724.2023.00031
“Both teacher and For the CaDDN [19] baseline, we used Ubuntu 18.04 OS, four NVIDIA A6000 GPUs, and trained for 80 epochs with a batch size of 8.”
MethodInfluentialSimulated Mental Imagery for Image Synthesis and its Application in Robotic Planning
Li, S. (2024) · eDiss (Georg-August-Universität Göttingen) · doi:10.53846/goediss-10515
An Omnidirectional Visual Perception Method Based on Parameter-Free Projection of Key Points
Feng, M., Chen, Y., Bai, H., & Wang, K. (2024) · International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) · doi:10.1109/ainit61980.2024.10581806
“CaDDN [3] and BEVDepth [4] are derived from LSS, but they incorporate depth labels to directly supervise the depth estimation.”
BackgroundToward Accurate Camera-based 3D Object Detection via Cascade Depth Estimation and Calibration
Wang, C., Qin, Y., Kang, Z., Ma, N., & Zhang, R. (2024) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra57147.2024.10610281
“CaDDN [7] learns a categorical depth distribution and then projects image features into BEV space.”
BackgroundMethodInfluentialWang, S., Gan, T., Fang, Y. Q., Xu, J., & Ling, Q. (2024) · Chinese Control Conference (CCC) · doi:10.23919/ccc63176.2024.10662789
“In the image branch, we follow OccDepth[15] and CADDN[18] to generate 3D image voxel features and subsequently convert them into BEV format.”
MethodPreBEV: Leveraging Predictive Flow for Enhanced Bird's-Eye View 3D Dynamic Object Detection
Ma, N., Li, S., Han, Y., & Wang, M. (2024) · IEEE International Conference on Software Quality, Reliability and Security Companion (QRS-C) · doi:10.1109/qrs-c63300.2024.00140
“CaDDN [8] is a 3D object detection method based on dense BEV features.”
MethodRePLAy: Remove Projective LiDAR Depthmap Artifacts via Exploiting Epipolar Geometry
Zhu, S., Ganesan, G. C., Kumar, A., & Liu, X. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73004-7_23
SFDet: Similar Fusion 3D Object Detection Using 3D Point Cloud and Image
Xu, Y., Xu, M., & Wang, Y. (2024) · International Conference on Intelligent Robotics and Control Engineering (IRCE) · doi:10.1109/irce62232.2024.10739796
“Reading et al. [11]estimated depth information for each pixel and projected the depth features into 3D space before using the bird’s eye view projection for 3D object detection.”
MethodSurfOcc: Surface-Based Feature Lifting for Vision-Centric 3D Occupancy Prediction
Ye, T., Gao, Z. F., Lin, Z., Liu, X.-Y., & Jin, R. (2024) · Lecture notes in computer science · doi:10.1007/978-981-96-0972-7_12
“These methods[36,16,26,28,32] usually follow the mechanism pioneered by Lift-Splat-Shoot[35] and improved it from different perspectives.”
InfluentialA Vision-Based Bird's Eye View Representation Network for 3D Objects in Open-Pit Mining Area
Tai, M., Liu, W., Wang, Z., Zhou, B., Yu, G., & Yang, S. (2024) · IEEE International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin58382.2024.10774405
Research on Adaptive 3D Object Detection Algorithm of Road Scene for Intelligent Driving
Lai, M. (2024) · IEEE International Conference on Civil Aviation Safety and Information Technology (ICCASIT) · doi:10.1109/iccasit62299.2024.10828017
“Finally, the features obtained above are input into the SSD network to complete the 3D target detection [7-8].”
Metadata
- Authors
- Reading, C., Harakeh, A., Chae, J., & Waslander, S. L.
- Affiliations
- University of Toronto (CA)
- Venue
- 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2021-06
- Pages
- 8551-8560
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W3133598367
- Cited by
- 595
- References
- 82
- Citations by year
- 2026: 57 · 2025: 143 · 2024: 173 · 2023: 136 · 2022: 73 · 2021: 13
- FWCI
- 32.30
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Vision and Imaging
- Keywords
- monocular 3D object detection, depth estimation, single-stage detectors, autonomous driving, Waymo Open Dataset, KITTI dataset
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
