Research map / Detection and segmentation / CNN features
Recurrent neural circuits for contour detection
Drew Linsley*, Junkyung Kim*, Alekh Ashok, Thomas Serre
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 |
|---|---|---|---|
| use recurrent horizontal and top-down fGRU connections | VGG16-based γ-net · BSDS500, reduced training sets | sample efficiency | Improved |
| use recurrent horizontal and top-down fGRU connections | U-Net-based γ-net · SNEMI3D and Ding, reduced training sets | sample efficiency | Improved |
| train γ-net rather than BDCN | BSDS500 · BSDS500 test | F1 ODS: 0.806 -> 0.802 | No clear effect |
| add recurrent horizontal and top-down fGRU connections | VGG16-based γ-net · BSDS500 | orientation-tilt illusion | Improved |
| correct the orientation-tilt illusion during training | γ-net · BSDS500 test | F1 ODS | Hurt |
| add horizontal-only connections with 15×15 kernels (ablation) | γ-net · BSDS500 | F1 ODS | Hurt |
| add horizontal-only connections with 3×3 kernels (ablation) | γ-net · BSDS500 | F1 ODS | Hurt |
| add top-down-only connections (ablation) | γ-net · BSDS500 | F1 ODS | Hurt |
| tie weights across recurrent timesteps (ablation) | γ-net · BSDS500 | F1 ODS | Improved |
| use recurrent processing rather than one timestep (ablation) | γ-net · BSDS500 | F1 ODS | Improved |
| add non-negativity constraints (ablation) | γ-net · BSDS500 | F1 ODS | Improved |
| add additive feedback parameters (ablation) | γ-net · BSDS500 | F1 ODS | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- BDCN — BDCN: Bi-Directional Cascade Network for Perceptual Edge Detection (TPAMI 2020) · better on 0 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
- Task-Driven Convolutional Recurrent Models of the Visual System (NeurIPS 2018 · similarity 0.85)
- Scene Categorization From Contours: Medial Axis Based Salience Measures (CVPR 2019 · similarity 0.85)
- Dynamic Refinement Network for Oriented and Densely Packed Object Detection (CVPR 2020 · similarity 0.84)
- Recurrent neural network dynamical systems for biological vision (NeurIPS 2024 · similarity 0.84)
- Hidden in Plain Sight: Evaluating Abstract Shape Recognition in Vision-Language Models (NeurIPS 2024 · similarity 0.84)
- Visual Pinwheel Centers Act as Geometric Saliency Detectors (NeurIPS 2024 · similarity 0.83)
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.
