Research map / Detection and segmentation / 3D detection and driving
Learning Object-Specific Distance From a Monocular Image
Jing Zhu, Yi Fang
Read the paperCheck 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 multiclass classifier (ablation) | base model (vgg16) · KITTI-distance | delta1 accuracy: 0.482 -> 0.587 | Improved |
| add multiclass classifier (ablation) | base model (vgg16) · KITTI-distance | Abs Rel: 0.658 -> 0.311 | Improved |
| add multiclass classifier (ablation) | enhanced model (vgg16) · KITTI-distance | delta1 accuracy: 0.486 -> 0.629 | Improved |
| add multiclass classifier (ablation) | enhanced model (vgg16) · KITTI-distance | Abs Rel: 0.541 -> 0.251 | Improved |
| add keypoint regressor and projection loss | base model (vgg16) · KITTI-distance (average) | delta1 accuracy: 0.587 -> 0.629 | Improved |
| add keypoint regressor and projection loss | base model (vgg16) · KITTI-distance (average) | Abs Rel: 0.311 -> 0.251 | Improved |
| add keypoint regressor and projection loss | base model (res50) · KITTI-distance (average) | delta1 accuracy: 0.503 -> 0.550 | Improved |
| add keypoint regressor and projection loss | base model (vgg16) · nuScenes (mini) distance | delta1 accuracy: 0.393 -> 0.535 | Improved |
| add keypoint regressor and projection loss | base model (res50) · nuScenes (mini) distance | Abs Rel: 0.466 -> 0.340 | Improved |
| use learned base model instead of IPM | IPM · KITTI-distance (average) | Abs Rel: 0.390 -> 0.311 | Improved |
| use VGG16 instead of ResNet50 backbone | base model · KITTI-distance (average) | delta1 accuracy: 0.503 -> 0.587 | Improved |
| use learned enhanced model instead of IPM | IPM · nuScenes (mini) distance | delta1 accuracy: 0.441 -> 0.535 | Improved |
Closest papers by meaning
- R4D: Utilizing Reference Objects for Long-Range Distance Estimation (ICLR 2022 · similarity 0.88)
- Geometry-Based Distance Decomposition for Monocular 3D Object Detection (ICCV 2021 · similarity 0.88)
- Depth-discriminative Metric Learning for Monocular 3D Object Detection (NeurIPS 2023 · similarity 0.87)
- Pseudo-LiDAR From Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving (CVPR 2019 · similarity 0.87)
- Wasserstein Distances for Stereo Disparity Estimation (NeurIPS 2020 · similarity 0.87)
- Monocular 3D Object Detection for Autonomous Driving (CVPR 2016 · similarity 0.87)
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.
