Research map / Detection and segmentation / CNN features
Straight to Shapes: Real-Time Detection of Encoded Shapes
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 |
|---|---|---|---|
| Regress to a learned shape embedding alongside locations and categories | YOLO-style detector · PASCAL SBD 2012 val | mAP^r_vol: 31.5 | Helped |
| Use a 20-dimensional rather than a 50-dimensional learned shape embedding (ablation) | YOLO-style shape detector · PASCAL SBD 2012 val | mAP^r_vol: 28.9 -> 31.5 | Helped |
| Use the learned embedding instead of downsampled binary masks (ablation) | YOLO-style shape detector · PASCAL SBD 2012 val | mAP^r_vol: 28.6 -> 31.5 | Helped |
| Use the learned embedding instead of radial vectors (ablation) | YOLO-style shape detector · PASCAL SBD 2012 val | mAP^r_vol: 29.0 -> 31.5 | Helped |
| Use a 20-dimensional rather than a 50-dimensional learned shape embedding (ablation) | YOLO-style shape detector · PASCAL SBD 2012 val | time (ms): 30.5 -> 28.0 | Helped |
| Use the learned shape encoding rather than downsampled masks (ablation) | Shape reconstruction representation · SBD validation set | average shape reconstruction error: 0.15 -> 0.125 | Helped |
| Use the learned shape encoding rather than radial vectors (ablation) | Shape reconstruction representation · SBD validation set | average shape reconstruction error: 0.13 -> 0.125 | Helped |
| Use the learned embedding rather than radial vectors (ablation) | Shape space · PASCAL Plus nearest-neighbour analysis | mA% majority class: 0.547 -> 0.800 | Helped |
| Use the learned embedding rather than radial vectors (ablation) | Shape space · PASCAL Plus nearest-neighbour analysis | mAV (azimuth): 1.163 -> 1.008 | Helped |
| Use the learned embedding rather than radial vectors (ablation) | Shape space · PASCAL Plus nearest-neighbour analysis | mAV (elevation): 0.294 -> 0.246 | Helped |
| Use the learned embedding rather than radial vectors (ablation) | Shape space · PASCAL Plus nearest-neighbour analysis | mAV (distance): 21.537 -> 17.968 | Helped |
Closest papers by meaning
- Shape or Texture: Understanding Discriminative Features in CNNs (ICLR 2021 · similarity 0.89)
- AShapeFormer: Semantics-Guided Object-Level Active Shape Encoding for 3D Object Detection via Transformers (CVPR 2023 · similarity 0.89)
- Object Contour Detection With a Fully Convolutional Encoder-Decoder Network (CVPR 2016 · similarity 0.88)
- RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder (NeurIPS 2020 · similarity 0.88)
- Visual Anagrams Reveal Hidden Differences in Holistic Shape Processing Across Vision Models (NeurIPS 2025 · similarity 0.87)
- Hidden in Plain Sight: Evaluating Abstract Shape Recognition in Vision-Language Models (NeurIPS 2024 · 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.