Research map / 3D vision / Depth and stereo
SpikeStereoNet: A Brain-Inspired Framework for Stereo Depth Estimation from Spike Streams
Zhuoheng Gao, Yihao Li, Jiyao Zhang, Rui Zhao, Tong Wu, Hao Tang, Zhaofei Yu, Hao Dong, Guozhang Chen, Tiejun Huang
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 recurrent connections and feedforward connections in RSNN (ablation) | SpikeStereoNet · Synthetic spike stereo test set | bad 2.0 (%): 12.07 -> 4.13 | Improved |
| add recurrent connections within layers (ablation) | SpikeStereoNet · Synthetic spike stereo test set | bad 2.0 (%): 11.05 -> 4.13 | Improved |
| add feedforward connections between layers (ablation) | SpikeStereoNet · Synthetic spike stereo test set | bad 2.0 (%): 5.86 -> 4.13 | Improved |
| add group normalization module (ablation) | SpikeStereoNet · Synthetic spike stereo test set | bad 2.0 (%): 7.49 -> 4.13 | Improved |
| add voltage regularization (ablation) | SpikeStereoNet · Synthetic spike stereo test set | bad 2.0 (%): 7.33 -> 4.13 | Improved |
| add firing-rate regularization (ablation) | SpikeStereoNet · Synthetic spike stereo test set | bad 2.0 (%): 6.38 -> 4.13 | Improved |
| add both regularizations (ablation) | SpikeStereoNet · Synthetic spike stereo test set | bad 2.0 (%): 7.77 -> 4.13 | Improved |
| use adaptive LIF instead of fixed LIF (ablation) | SpikeStereoNet RSNN · Synthetic spike stereo test set | bad 2.0 (%): 7.05 -> 4.13 | Improved |
| use ALIF spiking RSNN instead of GRU (ablation) | SpikeStereoNet RSNN · Synthetic spike stereo test set | bad 2.0 (%): 4.53 -> 4.13 | Improved |
| use ALIF spiking RSNN instead of LSTM (ablation) | SpikeStereoNet RSNN · Synthetic spike stereo test set | AvgErr (px): 0.49 -> 0.42 | Improved |
| use ALIF spiking RSNN instead of vanilla RNN (ablation) | SpikeStereoNet RSNN · Synthetic spike stereo test set | bad 2.0 (%): 7.28 -> 4.13 | Improved |
| use ALIF instead of raw SNN (ablation) | SpikeStereoNet RSNN · Synthetic spike stereo test set | bad 2.0 (%): 11.05 -> 4.13 | Improved |
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.
- MonSter — MonSter: Marry Monodepth to Stereo Unleashes Power (CVPR 2025) · better on 4 of 4 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
- Spike Transformer: Monocular Depth Estimation for Spiking Camera (ECCV 2022 · similarity 0.92)
- DSNeRF: Dynamic View Synthesis for Ultra-Fast Scenes From Continuous Spike Streams (TPAMI 2026 · similarity 0.88)
- SpikeNeRF: Learning Neural Radiance Fields from Continuous Spike Stream (CVPR 2024 · similarity 0.87)
- Finding Visual Saliency in Continuous Spike Stream (AAAI 2024 · similarity 0.86)
- Rethinking High-speed Image Reconstruction Framework with Spike Camera (AAAI 2025 · similarity 0.86)
- SpikeReveal: Unlocking Temporal Sequences from Real Blurry Inputs with Spike Streams (NeurIPS 2024 · similarity 0.85)
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.
