Research map / 3D vision / Depth and stereo
Sequential Adversarial Learning for Self-Supervised Deep Visual Odometry
Shunkai Li, Fei Xue, Xin Wang, Zike Yan, Hongbin Zha
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 frame-to-frame correlation code (ablation) | baseline framework · KITTI depth (50m cap) | Abs Rel: 0.218 -> 0.162 | Improved |
| add GAN discriminator on top of code (ablation) | framework with code · KITTI depth (50m cap) | Abs Rel: 0.162 -> 0.152 | Improved |
| add LSTM on top of code (ablation) | framework with code · KITTI depth (50m cap) | Abs Rel: 0.162 -> 0.148 | Improved |
| add LSTM to code+GAN (ablation) | framework with code+GAN · KITTI depth (50m cap) | Abs Rel: 0.152 -> 0.150 | No clear effect |
| add trajectory consistency loss (ablation) | code+GAN+LSTM · KITTI depth (50m cap) | Abs Rel: 0.150 -> 0.146 | Improved |
| add frame-to-frame correlation code (ablation) | baseline framework · KITTI odometry Seq.09 | ATE: 0.0072 -> 0.0069 | Improved |
| add GAN discriminator on top of code (ablation) | framework with code · KITTI odometry Seq.09 | ATE: 0.0069 -> 0.0064 | Improved |
| add LSTM on top of code (ablation) | framework with code · KITTI odometry Seq.09 | ATE: 0.0069 -> 0.0045 | Improved |
| add LSTM to code+GAN (ablation) | framework with code+GAN · KITTI odometry Seq.09 | ATE: 0.0064 -> 0.0036 | Improved |
| add trajectory consistency loss (ablation) | code+GAN+LSTM · KITTI odometry Seq.09 | ATE: 0.0036 -> 0.0030 | Improved |
| add trajectory consistency loss (ablation) | code+GAN+LSTM · KITTI odometry Seq.10 | ATE: 0.0036 -> 0.0029 | Improved |
| use proposed method instead of GeoNet | GeoNet · KITTI odometry Seq.09 | ATE: 0.012 -> 0.0030 | 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.
- GeoNet — GeoNet: Geometric Neural Network for Joint Depth and Surface Normal Estimation (CVPR 2018) · better on 5 of 5 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
- Learning Monocular Visual Odometry via Self-Supervised Long-Term Modeling (ECCV 2020 · similarity 0.91)
- Self-Supervised Deep Visual Odometry With Online Adaptation (CVPR 2020 · similarity 0.90)
- A Global Occlusion-Aware Approach to Self-Supervised Monocular Visual Odometry (AAAI 2021 · similarity 0.90)
- Generalizing to the Open World: Deep Visual Odometry With Online Adaptation (CVPR 2021 · similarity 0.89)
- STViT: Improving Self-Supervised Multi-Camera Depth Estimation with Spatial-Temporal Context and Adversarial Geometry Regularization (Student Abstract) (AAAI 2024 · similarity 0.85)
- Leveraging Consistent Spatio-Temporal Correspondence for Robust Visual Odometry (AAAI 2025 · 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.
