Research map / Video / Video segmentation and tracking
Lucid Data Dreaming for Video Object Segmentation
Anna Khoreva, Rodrigo Benenson, Eddy Ilg, Thomas Brox, Bernt Schiele
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 ImageNet pre-training (ablation) | LucidTracker- · DAVIS16 | mIoU: 82.0 -> 83.7 | Improved |
| add ImageNet pre-training (ablation) | LucidTracker- · SegTrack v2 | mIoU: 71.2 -> 76.8 | Improved |
| add per-video fine-tuning (ablation) | LucidTracker- · DAVIS16 | mIoU: 82.7 -> 83.7 | Improved |
| add per-video fine-tuning (ablation) | LucidTracker- · YouTubeObjects | mIoU: 72.3 -> 76.2 | Improved |
| add optical flow and mask warping (ablation) | LucidTracker- · SegTrack v2 | mIoU: 61.8 -> 76.8 | Improved |
| add optical flow and mask warping (ablation) | LucidTracker- · YouTubeObjects | mIoU: 74.7 -> 76.2 | Improved |
| add mask warping (ablation) | LucidTracker- · SegTrack v2 | mIoU: 70.5 -> 76.8 | Improved |
| use FlowNet2.0 instead of EpicFlow (ablation) | LucidTracker- · YouTubeObjects | mIoU: 71.3 -> 76.2 | Improved |
| use FlowNet2.0 instead of EpicFlow (ablation) | LucidTracker- · DAVIS16 | mIoU: 80.2 -> 83.7 | Improved |
| use default DenseCRF parameters instead of none (ablation) | LucidTracker- · SegTrack v2 | mIoU: 76.8 -> 72.2 | Hurt |
| tune DenseCRF parameters per dataset instead of default (ablation) | LucidTracker · SegTrack v2 | mIoU: 72.2 -> 77.6 | Improved |
| train on more videos (15 instead of 1) with first frames included (ablation) | RGB-only base model · DAVIS16 subset (15 videos) | mIoU: 78.3 -> 68.7 | Hurt |
Closest papers by meaning
- A Survey on Deep Learning Technique for Video Segmentation (TPAMI 2022 · similarity 0.90)
- Learning Video Object Segmentation With Visual Memory (ICCV 2017 · similarity 0.89)
- Pixel Objectness: Learning to Segment Generic Objects Automatically in Images and Videos (TPAMI 2018 · similarity 0.89)
- Video Object Segmentation Using Space-Time Memory Networks (ICCV 2019 · similarity 0.89)
- From ViT Features to Training-free Video Object Segmentation via Streaming-data Mixture Models (NeurIPS 2023 · similarity 0.88)
- LiVOS: Light Video Object Segmentation with Gated Linear Matching (CVPR 2025 · 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.
