Research map / Video / Video segmentation and tracking
Video Scene Parsing With Predictive Feature Learning
Xiaojie Jin, Xin Li, Huaxin Xiao, Xiaohui Shen, Zhe Lin, Jimei Yang, Yunpeng Chen, Jian Dong, Luoqi Liu, Zequn Jie, Jiashi Feng, Shuicheng Yan
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 FPNet frozen spatiotemporal features to the image parser (ablation) | VGG16 IPNet · Cityscapes val | mIoU: 63.4 -> 68.6 | Improved |
| use PPNet features instead of FPNet features (ablation) | VGG16 IPNet · Cityscapes val | mIoU: +0.5 (68.6 -> 69.1) | Improved |
| train FPNet unsupervised from random init (VGG11) rather than ImageNet-pretrained VGG16 (ablation) | FPNet · Cityscapes val | mIoU: 68.4 vs 68.6 | No clear effect |
| joint end-to-end training of PPNet and IPNet instead of frozen features (ablation) | PEARL VGG16 · Cityscapes val | mIoU: 69.1 -> 69.8 | Improved |
| replace optical flow input channels with PEARL predictive features (ablation) | VGG16 parser · Cityscapes val | mIoU: 64.5 -> 69.8 | Improved |
| replace optical flow input channels with PEARL predictive features (ablation) | Res101 parser · Cityscapes val | mIoU: 72.7 -> 74.9 | Improved |
| add optical flow channels to input (ablation) | VGG16 baseline · Cityscapes val | mIoU: 63.4 -> 64.5 | Improved |
| add optical flow channels to input (ablation) | Res101 baseline · Cityscapes val | mIoU: 72.5 -> 72.7 | No clear effect |
| add AdapNet (one 1x1 conv + ReLU) (ablation) | PEARL VGG16 · Cityscapes val | mIoU: +1.1 | Improved |
| add AdapNet (one 1x1 conv + ReLU) (ablation) | PEARL Res101 · Cityscapes val | mIoU: +0.3 | Improved |
| deeper AdapNet with more conv layers (ablation) | PEARL · Cityscapes val | mIoU: marginal improvements | No clear effect |
| apply PEARL to baseline | VGG16 · Cityscapes val | mIoU: 63.4 -> 69.8 (+6.4) | Improved |
Closest papers by meaning
- VSPW: A Large-scale Dataset for Video Scene Parsing in the Wild (CVPR 2021 · similarity 0.91)
- Temporal Pixel-Level Semantic Understanding Through the VSPW Dataset (TPAMI 2023 · similarity 0.90)
- Scene Parsing With Global Context Embedding (ICCV 2017 · similarity 0.89)
- Predicting Scene Parsing and Motion Dynamics in the Future (NeurIPS 2017 · similarity 0.89)
- Adaptive Context Network for Scene Parsing (ICCV 2019 · similarity 0.87)
- Semantic Flow for Fast and Accurate Scene Parsing (ECCV 2020 · 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.
