Research map / Detection and segmentation / CNN features
Accurate Single Stage Detector Using Recurrent Rolling Convolution
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 recurrent rolling convolution for contextual feature aggregation | SSD with reduced VGG-16 · KITTI validation car, IoU 0.8 | mAP: 76.12% -> 82.82% | Helped |
| Add recurrent rolling convolution for contextual feature aggregation | Single-stage detector · KITTI car test, hard | AP: 85.66% -> 86.97% | Helped |
| Add recurrent rolling convolution for contextual feature aggregation | Single-stage detector · KITTI pedestrian test, moderate | AP: 73.70% -> 75.33% | Helped |
| Add recurrent rolling convolution for contextual feature aggregation | Single-stage detector · KITTI cyclist test, moderate | AP: 75.46% -> 76.47% | Helped |
| Use NMS over outputs 3 through 5 rather than outputs 2 through 6 (ablation) | RRC · KITTI validation car | mAP: 90.32% -> 90.65%; 89.01% -> 89.27%; 82.19% -> 82.82% | Helped |
| Add recurrent rolling convolution iterations (ablation) | RRC · KITTI validation set | average loss: 1.461 -> 1.374 -> 1.357 -> 1.361 -> 1.366 -> 1.375 | Mixed |
Closest papers by meaning
- CIA-SSD: Confident IoU-Aware Single-Stage Object Detector From Point Cloud (AAAI 2021 · similarity 0.86)
- R-FCN: Object Detection via Region-based Fully Convolutional Networks (NeurIPS 2016 · similarity 0.86)
- Improving Occlusion and Hard Negative Handling for Single-Stage Pedestrian Detectors (CVPR 2018 · similarity 0.85)
- Feature Pyramid Networks for Object Detection (CVPR 2017 · similarity 0.85)
- Object Detection With Location-Aware Deformable Convolution and Backward Attention Filtering (CVPR 2019 · similarity 0.85)
- Cascade R-CNN: Delving Into High Quality Object Detection (CVPR 2018 · similarity 0.84)
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.