AI Paper Atlas

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

CVPR 2020 · Detection and segmentation › 3D detection and driving · approach: Unified LiDAR 3D detection · 19 citations (OpenAlex)

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 changedWhereMeasuredEffect
Add cylindrical view-specific kernelsFaster R-CNN with FPN · COCO minivalAP: 40.7 -> 41.8Improved
Add cylindrical view-specific kernelsFaster R-CNN with FPN · COCO minivalAR: 54.1 -> 56.3Improved
Add CCNs and semi-supervised trainingFaster R-CNN with FPN · Pascal 3D+ valmAP: 71.4 -> 73.7Improved
Add CCNs and semi-supervised trainingFaster R-CNN with FPN · Pascal 3D+ valmAVP24: 44.4 -> 46.6Improved
Add CCNsFaster R-CNN · KITTI val carsAP Easy: 81.74 -> 86.17Improved
Add CCNsFaster R-CNN · KITTI val carsAOS Easy: 79.46 -> 85.01Improved
Replace a view-agnostic kernel with cylindrical view-specific kernels (ablation)ResNet-101 joint category and viewpoint estimation · Pascal 3D+ ground-truth boxestop-1: 0.91 -> 0.95Improved
Replace a view-agnostic kernel with cylindrical view-specific kernels (ablation)ResNet-101 joint category and viewpoint estimation · Pascal 3D+ ground-truth boxesAccπ/6: 0.56 -> 0.66Improved
Use 18 viewpoint bins instead of 24 (ablation)CCNs · Pascal 3D+ ground-truth boxesAccπ/6: 0.66 -> 0.63Hurt
Use 30 viewpoint bins instead of 24 (ablation)CCNs · Pascal 3D+ ground-truth boxesAccπ/6: 0.66 -> 0.63Hurt
Use 18 viewpoint bins instead of 24 (ablation)CCNs · Pascal 3D+ ground-truth boxesMedErr: 15.5 -> 17.3Hurt
Use 30 viewpoint bins instead of 24 (ablation)CCNs · Pascal 3D+ ground-truth boxesMedErr: 15.5 -> 17.7Hurt

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy