Research map / Detection and segmentation / CNN features
End-to-End High Dynamic Range Camera Pipeline Optimization
Nicolas Robidoux, Luis E. Garcia Capel, Dong-eun Seo, Avinash Sharma, Federico Ariza, Felix Heide
Check 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 |
|---|---|---|---|
| Jointly optimize sensor, ISP, and detector weights instead of using expert-tuned settings | HDR camera pipeline with YOLOv4 · Automotive object detection test set | mAP (IoU>0.5): 0.250 -> 0.584 | Improved |
| Jointly optimize sensor, ISP, and detector weights instead of using Mosleh et al. optimization | HDR camera pipeline with YOLOv4 · Automotive object detection test set | mAP (IoU>0.5): 0.367 -> 0.584 | Improved |
| Jointly optimize sensor, ISP, and detector weights instead of using expert-tuned settings | HDR camera pipeline with YOLOv4 · Automotive object detection test set | mAP (IoU>0.75): 0.244 -> 0.561 | Improved |
| Jointly optimize sensor, ISP, and detector weights instead of using Mosleh et al. optimization | HDR camera pipeline with YOLOv4 · Automotive object detection test set | mAP (IoU>0.75): 0.356 -> 0.561 | Improved |
| Jointly optimize sensor, ISP, and detector weights instead of using expert-tuned settings | HDR camera pipeline with YOLOv4 · Automotive object detection test set | mAR: 0.235 -> 0.560 | Improved |
| Jointly optimize sensor, ISP, and detector weights instead of using Mosleh et al. optimization | HDR camera pipeline with YOLOv4 · Automotive object detection test set | mAR: 0.352 -> 0.560 | Improved |
| Use the proposed perceptual optimization rather than the combined Mosleh et al. and expert-tuned settings | AR0231AT + AP0202AT HDR pipeline · HDR perceptual optimization lab | FSITM mean: 0.335 -> 0.333 | Improved |
| Use the proposed perceptual optimization rather than the combined Mosleh et al. and expert-tuned settings | AR0231AT + AP0202AT HDR pipeline · HDR perceptual optimization lab | CWLP mean: 4.083 -> 4.045 | Improved |
| Use the proposed perceptual optimization rather than the combined Mosleh et al. and expert-tuned settings | AR0231AT + AP0202AT HDR pipeline · HDR perceptual optimization lab | Zippering mean = worst: 0.068 -> 0.071 | Hurt |
| Use the proposed perceptual optimization rather than the combined Mosleh et al. and expert-tuned settings | AR0231AT + AP0202AT HDR pipeline · HDR perceptual optimization lab | SNR worst: 23.60 -> 24.89 | Improved |
| Continue joint optimization from one iteration to convergence (ablation) | Proposed HDR camera pipeline with YOLOv4 · Automotive object detection test set | mAP (IoU>0.5): 0.563 -> 0.584 | Improved |
| Continue joint optimization from one iteration to convergence (ablation) | Proposed HDR camera pipeline with YOLOv4 · Automotive object detection test set | mAP (IoU>0.75): 0.540 -> 0.561 | Improved |
Closest papers by meaning
- Real-Time Scene-Adaptive Tone Mapping for High-Dynamic Range Object Detection (NeurIPS 2025 · similarity 0.89)
- Neural Auto-Exposure for High-Dynamic Range Object Detection (CVPR 2021 · similarity 0.89)
- HDRFlow: Real-Time HDR Video Reconstruction with Large Motions (CVPR 2024 · similarity 0.88)
- Neural Exposure Fusion for High-Dynamic Range Object Detection (CVPR 2024 · similarity 0.88)
- Towards Real-World HDR Video Reconstruction: A Large-Scale Benchmark Dataset and A Two-Stage Alignment Network (CVPR 2024 · similarity 0.88)
- Self-Supervised High Dynamic Range Imaging with Multi-Exposure Images in Dynamic Scenes (ICLR 2024 · 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.
