Latency-aware Road Anomaly Segmentation in Videos: A Photorealistic Dataset and New Metrics
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913191383007232 |
|---|---|
| author | Tian, Beiwen Gao, Huan-ang Cui, Leiyao Zheng, Yupeng Luo, Lan Wang, Baofeng Zhi, Rong Zhou, Guyue Zhao, Hao |
| author_facet | Tian, Beiwen Gao, Huan-ang Cui, Leiyao Zheng, Yupeng Luo, Lan Wang, Baofeng Zhi, Rong Zhou, Guyue Zhao, Hao |
| contents | In the past several years, road anomaly segmentation is actively explored in the academia and drawing growing attention in the industry. The rationale behind is straightforward: if the autonomous car can brake before hitting an anomalous object, safety is promoted. However, this rationale naturally calls for a temporally informed setting while existing methods and benchmarks are designed in an unrealistic frame-wise manner. To bridge this gap, we contribute the first video anomaly segmentation dataset for autonomous driving. Since placing various anomalous objects on busy roads and annotating them in every frame are dangerous and expensive, we resort to synthetic data. To improve the relevance of this synthetic dataset to real-world applications, we train a generative adversarial network conditioned on rendering G-buffers for photorealism enhancement. Our dataset consists of 120,000 high-resolution frames at a 60 FPS framerate, as recorded in 7 different towns. As an initial benchmarking, we provide baselines using latest supervised and unsupervised road anomaly segmentation methods. Apart from conventional ones, we focus on two new metrics: temporal consistency and latencyaware streaming accuracy. We believe the latter is valuable as it measures whether an anomaly segmentation algorithm can truly prevent a car from crashing in a temporally informed setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_04942 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Latency-aware Road Anomaly Segmentation in Videos: A Photorealistic Dataset and New Metrics Tian, Beiwen Gao, Huan-ang Cui, Leiyao Zheng, Yupeng Luo, Lan Wang, Baofeng Zhi, Rong Zhou, Guyue Zhao, Hao Computer Vision and Pattern Recognition In the past several years, road anomaly segmentation is actively explored in the academia and drawing growing attention in the industry. The rationale behind is straightforward: if the autonomous car can brake before hitting an anomalous object, safety is promoted. However, this rationale naturally calls for a temporally informed setting while existing methods and benchmarks are designed in an unrealistic frame-wise manner. To bridge this gap, we contribute the first video anomaly segmentation dataset for autonomous driving. Since placing various anomalous objects on busy roads and annotating them in every frame are dangerous and expensive, we resort to synthetic data. To improve the relevance of this synthetic dataset to real-world applications, we train a generative adversarial network conditioned on rendering G-buffers for photorealism enhancement. Our dataset consists of 120,000 high-resolution frames at a 60 FPS framerate, as recorded in 7 different towns. As an initial benchmarking, we provide baselines using latest supervised and unsupervised road anomaly segmentation methods. Apart from conventional ones, we focus on two new metrics: temporal consistency and latencyaware streaming accuracy. We believe the latter is valuable as it measures whether an anomaly segmentation algorithm can truly prevent a car from crashing in a temporally informed setting. |
| title | Latency-aware Road Anomaly Segmentation in Videos: A Photorealistic Dataset and New Metrics |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2401.04942 |