Research map / Detection and segmentation / Open-set and few-shot detection
TRoVE: Transforming Road Scene Datasets into Photorealistic Virtual Environments
Shubham Dokania, Anbumani Subramanian, Manmohan Chandraker, C.V. Jawahar
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 synthetic data mixed into batches with real data | DeepLabV2 R-50 · Cityscapes (full real) | mIoU: 70.25 -> 70.82 | Improved |
| add color-adjusted synthetic data mixed with full real data | DeepLabV2 R-50 · Cityscapes (full real) | mIoU: 70.25 -> 71.98 | Improved |
| add synthetic data mixed with partial real data | DeepLabV2 R-50 · Cityscapes (1000 real) | mIoU: 61.44 -> 67.21 | Improved |
| add synthetic data mixed with full real data | DeepLabV2 R-50 · KITTI-STEP (full real) | mIoU: 59.81 -> 65.37 | Improved |
| add synthetic data mixed with partial real data | DeepLabV2 R-50 · KITTI-STEP (1000 real) | mIoU: 56.44 -> 61.72 | Improved |
| mix synthetic and real in batch instead of synthetic pretraining then fine-tuning (ablation) | S + R · Cityscapes (full real) | mIoU: 70.23 -> 70.82 | Improved |
| mix synthetic and real in batch instead of fine-tuning (ablation) | S + P · Cityscapes (1000 real) | mIoU: 63.56 -> 67.21 | Improved |
| mix synthetic and real in batch instead of fine-tuning (ablation) | S + R · KITTI-STEP (full real) | mIoU: 59.79 -> 65.37 | Improved |
| add color transformation to synthetic data, training on synthetic only (ablation) | S · Cityscapes | mIoU: 30.63 -> 37.03 | Improved |
| add color transformation to synthetic data, training on synthetic only (ablation) | S · KITTI-STEP | mIoU: 27.64 -> 46.14 | Improved |
| add color transformation to synthetic data mixed with partial real (ablation) | S + P [M] · Cityscapes (1000 real) | mIoU: 67.21 -> 65.75 | Hurt |
| add color transformation to synthetic data mixed with partial real (ablation) | S + P [M] · KITTI-STEP (1000 real) | mIoU: 61.72 -> 63.09 | Improved |
Closest papers by meaning
- Toward Generating Realistic 3D Semantic Training Data for Autonomous Driving (TPAMI 2026 · similarity 0.89)
- MAPLM: A Real-World Large-Scale Vision-Language Benchmark for Map and Traffic Scene Understanding (CVPR 2024 · similarity 0.86)
- SceneCrafter: Controllable Multi-View Driving Scene Editing (CVPR 2025 · similarity 0.86)
- Weakly Supervised Semantic Segmentation for Driving Scenes (AAAI 2024 · similarity 0.86)
- RoadSceneBench: A Lightweight Benchmark for Mid-Level Road Scene Understanding (CVPR 2026 · similarity 0.85)
- WinSyn: : A High Resolution Testbed for Synthetic Data (CVPR 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.
