Research map / Video / Video segmentation and tracking
TMCOSS: Thresholded Multi-Criteria Online Subset Selection for Data-Efficient Autonomous Driving
Soumi Das, Harikrishna Patibandla, Suparna Bhattacharya, Kshounis Bera, Niloy Ganguly, Sourangshu Bhattacharya
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 |
|---|---|---|---|
| replace additive loss term with thresholded concave term (TMCOSS-TL vs MCOSS) | MCOSS · CARLA One-Turn, training conditions, 100:20 | episodes completed: 8 -> 8, test 7 -> 9 | Improved |
| add bucket-specific relative angle loss (BL) instead of total loss (ablation) | TMCOSS-TL · CARLA One-Turn, training conditions, 100:20 | episodes completed: 8 -> 10 | Improved |
| add bucket-specific relative angle loss (BL) instead of total loss (ablation) | TMCOSS-TL · CARLA One-Turn, training conditions, 100:7 | episodes completed: 7 -> 8 | Improved |
| use TMCOSS-BL instead of MCOSS | MCOSS · CARLA One-Turn, training conditions, 100:7 | episodes completed: 5 -> 8 | Improved |
| use TMCOSS-BL instead of SubMCOSS | SubMCOSS · CARLA One-Turn, training conditions, 100:7 | episodes completed: 3 -> 8 | Improved |
| add pointwise loss criterion additively (MCOSS vs OSS) | OSS · CARLA One-Turn, test conditions, 100:20 | episodes completed: 7 -> 7 | No clear effect |
| use TMCOSS-BL instead of TMCOSS-TL (ablation) | TMCOSS-TL · CARLA affordances, 100:20 | Red Light accuracy: 96.05 -> 92.83 | Hurt |
| use TMCOSS instead of MCOSS | MCOSS · BDD drivable area segmentation, 100:20 | Drivable Area IoU: 76.0 -> 80.0 | Improved |
| use TMCOSS instead of MCOSS | MCOSS · BDD drivable area segmentation, 100:20 | Alternate Drive Area IoU: 59.0 -> 65.0 | Improved |
| use TMCOSS instead of MCOSS | MCOSS · Cityscapes semantic segmentation, 100:20 | Mean IoU: 72.33 -> 79.33 | Improved |
| use TMCOSS instead of MCOSS | MCOSS · Cityscapes semantic segmentation, 100:20 | Wall IoU: 29.0 -> 50.0 | Improved |
| use uniform sampling instead of OSS | OSS · CARLA One-Turn, training conditions, 100:20 | episodes completed: 7 -> 3 | Hurt |
Closest papers by meaning
- Panacea: Panoramic and Controllable Video Generation for Autonomous Driving (CVPR 2024 · similarity 0.85)
- Prune2Drive: A Plug-and-Play Framework for Accelerating Vision-Language Models in Autonomous Driving (CVPR 2026 · similarity 0.85)
- DriveGEN: Generalized and Robust 3D Detection in Driving via Controllable Text-to-Image Diffusion Generation (CVPR 2025 · similarity 0.85)
- Glad: A Streaming Scene Generator for Autonomous Driving (ICLR 2025 · similarity 0.85)
- Generalized Predictive Model for Autonomous Driving (CVPR 2024 · similarity 0.85)
- H-MBA: Hierarchical MamBa Adaptation for Multi-Modal Video Understanding in Autonomous Driving (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.
