Research map / Detection and segmentation / 3D detection and driving
Cylindrical Convolutional Networks for Joint Object Detection and Viewpoint Estimation
Sunghun Joung, Seungryong Kim, Hanjae Kim, Minsu Kim, Ig-Jae Kim, Junghyun Cho, Kwanghoon Sohn
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add cylindrical view-specific kernels | Faster R-CNN with FPN · COCO minival | AP: 40.7 -> 41.8 | Improved |
| Add cylindrical view-specific kernels | Faster R-CNN with FPN · COCO minival | AR: 54.1 -> 56.3 | Improved |
| Add CCNs and semi-supervised training | Faster R-CNN with FPN · Pascal 3D+ val | mAP: 71.4 -> 73.7 | Improved |
| Add CCNs and semi-supervised training | Faster R-CNN with FPN · Pascal 3D+ val | mAVP24: 44.4 -> 46.6 | Improved |
| Add CCNs | Faster R-CNN · KITTI val cars | AP Easy: 81.74 -> 86.17 | Improved |
| Add CCNs | Faster R-CNN · KITTI val cars | AOS Easy: 79.46 -> 85.01 | Improved |
| Replace a view-agnostic kernel with cylindrical view-specific kernels (ablation) | ResNet-101 joint category and viewpoint estimation · Pascal 3D+ ground-truth boxes | top-1: 0.91 -> 0.95 | Improved |
| Replace a view-agnostic kernel with cylindrical view-specific kernels (ablation) | ResNet-101 joint category and viewpoint estimation · Pascal 3D+ ground-truth boxes | Accπ/6: 0.56 -> 0.66 | Improved |
| Use 18 viewpoint bins instead of 24 (ablation) | CCNs · Pascal 3D+ ground-truth boxes | Accπ/6: 0.66 -> 0.63 | Hurt |
| Use 30 viewpoint bins instead of 24 (ablation) | CCNs · Pascal 3D+ ground-truth boxes | Accπ/6: 0.66 -> 0.63 | Hurt |
| Use 18 viewpoint bins instead of 24 (ablation) | CCNs · Pascal 3D+ ground-truth boxes | MedErr: 15.5 -> 17.3 | Hurt |
| Use 30 viewpoint bins instead of 24 (ablation) | CCNs · Pascal 3D+ ground-truth boxes | MedErr: 15.5 -> 17.7 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Faster R-CNN — Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (TPAMI 2016) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- RotationNet: Joint Object Categorization and Pose Estimation Using Multiviews From Unsupervised Viewpoints (CVPR 2018 · similarity 0.87)
- Learning Spherical Convolution for Fast Features from 360° Imagery (NeurIPS 2017 · similarity 0.86)
- RotationNet for Joint Object Categorization and Unsupervised Pose Estimation from Multi-View Images (TPAMI 2019 · similarity 0.86)
- A Comparative Analysis and Study of Multiview CNN Models for Joint Object Categorization and Pose Estimation (ICML 2016 · similarity 0.86)
- To the Point: Efficient 3D Object Detection in the Range Image With Graph Convolution Kernels (CVPR 2021 · similarity 0.85)
- R-FCN: Object Detection via Region-based Fully Convolutional Networks (NeurIPS 2016 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
