Deep Supervision with Shape Concepts for Occlusion-Aware 3D Object Parsing
The DOI matches its Crossref record; 7 of the 7 references it cites were found in a registry or an index.
Abstract
Monocular 3D object parsing is highly desirable in various scenarios including occlusion reasoning and holistic scene interpretation. We present a deep convolutional neural network (CNN) architecture to localize semantic parts in 2D image and 3D space while inferring their visibility states, given a single RGB image. Our key insight is to exploit domain knowledge to regularize the network by deeply supervising its hidden layers, in order to sequentially infer intermediate concepts associated with the final task. To acquire training data in desired quantities with ground truth 3D shape and relevant concepts, we render 3D object CAD models to generate large-scale synthetic data and simulate challenging occlusion configurations between objects. We train the network only on synthetic data and demonstrate state-of-the-art performances on real image benchmarks including an extended version of KITTI, PASCAL VOC, PASCAL3D+ and IKEA for 2D and 3D keypoint localization and instance segmentation. The empirical results substantiate the utility of our deep supervision scheme by demonstrating effective transfer of knowledge from synthetic data to real images, resulting in less overfitting compared to standard end-to-end training.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
ShapeNet: An Information-Rich 3D Model Repository
Chang, A. X., Funkhouser, T., Guibas, L., Hanrahan, P., Huang, Q., Li, Z., Savarese, S., Savva, M., Song, S., Su, H., et al. (2015) · arXiv · doi:10.48550/arxiv.1512.03012 · Cited by 2,330
Playing for Data: Ground Truth from Computer Games
Richter, S. R., Vineet, V., Roth, S., & Koltun, V. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46475-6_7 · Cited by 1,815
Learning to generate chairs with convolutional neural networks
Dosovitskiy, A., Springenberg, J. T., & Brox, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298761 · Cited by 640
Multi-view 3D Models from Single Images with a Convolutional Network
Tatarchenko, M., Dosovitskiy, A., & Brox, T. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46478-7_20 · Cited by 337 · cited 3×
“However, methods such as [34, 25] are quantitatively evaluated only on synthetic data and seem to achieve limited generalization to real images.”
BackgroundMethodParsing IKEA Objects: Fine Pose Estimation
Lim, J. J., Pirsiavash, H., & Torralba, A. (2013) · 2013 IEEE International Conference on Computer Vision · doi:10.1109/iccv.2013.372 · Cited by 249 · cited 3×
“In Section 4, we quantitatively demonstrate significant improvements over prior state-of-the-art for 2D keypoint and 3D structure prediction on PASCAL VOC, PASCAL3D+[40], IKEA[19] and an extended KITTI [6] dataset (KITTI-3D).”
MethodMarr Revisited: 2D-3D Alignment via Surface Normal Prediction
Bansal, A., Russell, B., & Gupta, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.642 · Cited by 213
Research works printed but not linked: 38
- 10.1109/CVPR.2015.7298800 · DOI · Paper page
- 10.1109/TPAMI.2015.2497699 · DOI · Paper page
- 10.1109/ICCV.2015.308 · DOI · Paper page
- smith, 1986, Perception of Organization in a Random Stimulus
- simonyan, 2014, Very Deep Convolutional Networks for Large-scale Image Recognition
- 10.1109/34.857006 · DOI · Paper page
- 10.1007/978-3-319-46466-4_22 · DOI
- 10.1109/CVPR.2015.7298758 · DOI · Paper page
- torresani, 2003, Learning nonrigid 3d shape from 2d motion, Advances in neural information processing systems
- ioffe, 2015, Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift, JMLR
- 10.1109/WACV.2014.6836101 · DOI · Paper page
- jia, 2014, Caffe Convolutional Architecture for Fast Feature Embedding
- 10.1109/CVPR.2016.354 · DOI · Paper page
- 10.1109/CVPR.2015.7298807 · DOI · Paper page
- kulkarni, 2015, Deep convolutional inverse graphics network, NIPS
- lee, 2015, Deeply-Supervised Nets, AISTATS
- lee, 1985, Determination of 3D human body postures from a single view, CVGIP
- 10.1007/978-3-319-10599-4_31 · DOI
- rezende, 2016, Unsupervised learning of 3d structure from images, NIPS
- choy, 2016, 3DR2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction, ECCV
- pepik, 2013, Occlusion patterns for object class detection, CVPR
- 10.1109/CVPR.2012.6248074 · DOI · Paper page
- glorot, 2010, Understanding the difficulty of training deep feedforward neural networks, AISTATS
- 10.1109/CVPR.2014.81 · DOI · Paper page
- gupta, 2015, Inferring 3d object pose in rgb-d images
- 10.1109/CVPR.2014.487 · DOI · Paper page
- long, 2014, Do convnets learn correspondence?, NIPS
- 10.1007/s11263-014-0780-y · DOI
- marr, 1982, Vision
- 10.1007/978-1-4613-2551-2 · DOI · Paper page
- 10.1007/978-3-319-46454-1_4 · DOI
- 10.1109/34.42852 · DOI · Paper page
- 10.1109/CVPR.2011.5995741 · DOI · Paper page
- massa, 2015, Deep Exemplar 2D-3D Detection by Adapting from Real to Rendered Views, CVPR
- zia, 2009, Acquisition of a Dense 3D Model Database for Robotic Vision, ICAR
- 10.1109/CVPR.2015.7298639 · DOI · Paper page
- 10.1109/CVPR.2016.20 · DOI · Paper page
- moreno, 2016, Overcoming occlusion with inverse graphics, ECCV
Cited by
Image-Based 3D Object Reconstruction: State-of-the-Art and Trends in the Deep Learning Era
Han, X.-F., Laga, H., & Bennamoun, M. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2954885 · Cited by 431
GS3D: An Efficient 3D Object Detection Framework for Autonomous Driving
Li, B., Ouyang, W., Sheng, L., Zeng, X., & Wang, X. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00111 · Cited by 361
Multi-task Self-Supervised Learning for Human Activity Detection
Saeed, A., Ozcelebi, T., & Lukkien, J. (2019) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3328932 · Cited by 346
“ese techniques have achieved indisputable empirical success across a broad spectrum of problems [4, 21, 28, 35, 43, 56, 65, 66] thanks to the increasing dataset sizes and computing power availability.”
Background3D-RCNN: Instance-Level 3D Object Reconstruction via Render-and-Compare
Kundu, A., Li, Y., & Rehg, J. M. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00375 · Cited by 328
Cheng, D., Zhang, L., Bu, C., Wu, H., & Song, A. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110789 · Cited by 55
Chen, S., Peng, C., Yang, B., Lin, J., Zhou, L., Jiang, Z., Liu, Z., Liu, Y., & Tang, L. (2026) · npj Artificial Intelligence · doi:10.1038/s44387-026-00083-5 · Cited by 9
Metadata
- Authors
- Li, C., Zia, M. Z., Tran, Q.-H., Yu, X., Hager, G. D., & Chandraker, M.
- Venue
- 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2017-07
- Pages
- 388-397
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W2964057995
- Cited by
- 82
- References
- 7
- FWCI
- 3.72
- Type
- conference paper
- Language
- en
- Open access
- closed
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.