Abstract
In this paper, we propose our method for vehicle detection with multiple adaptive vehicle detectors and velocity estimation with landmark-based scanlines. Inspired by the idea for tiny object detection, we use Faster R-CNN with Resnet-101 to create different specialized vehicle detectors corresponding to different levels of details and poses. We propose a heuristic to check the fitness of a particular vehicle detector to a specific region in camera's view by the mean velocity direction and the mean object size. By this way, we can determine an adaptive set of appropriate vehicle detectors for each region in camera's view. Thus our system is expected to detect vehicles with high accuracy, both in precision and recall, even with tiny objects. We exploit the U.S. road rules for the length and distance of broken white lines on roads to propose our method for vehicle's velocity estimation using such landmarks. We determine equally-distributed scanlines, virtual parallel lines that are nearly-perpendicular to the road direction, with reference to the line connecting the corresponding ends of multiple broken white lines. From the timespan for a vehicle to cross two consecutive virtual scanlines, we can calculate the average vehicle's velocity within that road segment. We also refine the speed estimation by detecting when a vehicle stops at a traffic light, and smooth the results with a moving average filter. Experiments on the dataset of Traffic Flow Analysis from NVIDIA AI City Challenge 2018 show that our method achieves the perfect detect rate of 100%, the average velocity difference of 6.9762 mph on freeways, and 8.9144 mph on both freeways and urban roads.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018 |
| Publisher | IEEE Computer Society |
| Pages | 100-107 |
| Number of pages | 8 |
| ISBN (Electronic) | 9781538661000 |
| DOIs | |
| State | Published - Dec 13 2018 |
| Event | 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018 - Salt Lake City, United States Duration: Jun 18 2018 → Jun 22 2018 |
Publication series
| Name | IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops |
|---|---|
| Volume | 2018-June |
| ISSN (Print) | 2160-7508 |
| ISSN (Electronic) | 2160-7516 |
Other
| Other | 31st Meeting of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018 |
|---|---|
| Country | United States |
| City | Salt Lake City |
| Period | 6/18/18 → 6/22/18 |
Fingerprint
ASJC Scopus subject areas
- Computer Vision and Pattern Recognition
- Electrical and Electronic Engineering
Cite this
Traffic flow analysis with multiple adaptive vehicle detectors and velocity estimation with landmark-based scanlines. / Tran, Minh Triet; Dinh-Duy, Tung; Truong, Thanh Dat; Ton-That, Vinh; Do, Thanh Nhon; Luong, Quoc An; Nguyen, Thanh An; Nguyen, Vinh Tiep; Do, Minh N.
Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018. IEEE Computer Society, 2018. p. 100-107 8575374 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops; Vol. 2018-June).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Traffic flow analysis with multiple adaptive vehicle detectors and velocity estimation with landmark-based scanlines
AU - Tran, Minh Triet
AU - Dinh-Duy, Tung
AU - Truong, Thanh Dat
AU - Ton-That, Vinh
AU - Do, Thanh Nhon
AU - Luong, Quoc An
AU - Nguyen, Thanh An
AU - Nguyen, Vinh Tiep
AU - Do, Minh N.
PY - 2018/12/13
Y1 - 2018/12/13
N2 - In this paper, we propose our method for vehicle detection with multiple adaptive vehicle detectors and velocity estimation with landmark-based scanlines. Inspired by the idea for tiny object detection, we use Faster R-CNN with Resnet-101 to create different specialized vehicle detectors corresponding to different levels of details and poses. We propose a heuristic to check the fitness of a particular vehicle detector to a specific region in camera's view by the mean velocity direction and the mean object size. By this way, we can determine an adaptive set of appropriate vehicle detectors for each region in camera's view. Thus our system is expected to detect vehicles with high accuracy, both in precision and recall, even with tiny objects. We exploit the U.S. road rules for the length and distance of broken white lines on roads to propose our method for vehicle's velocity estimation using such landmarks. We determine equally-distributed scanlines, virtual parallel lines that are nearly-perpendicular to the road direction, with reference to the line connecting the corresponding ends of multiple broken white lines. From the timespan for a vehicle to cross two consecutive virtual scanlines, we can calculate the average vehicle's velocity within that road segment. We also refine the speed estimation by detecting when a vehicle stops at a traffic light, and smooth the results with a moving average filter. Experiments on the dataset of Traffic Flow Analysis from NVIDIA AI City Challenge 2018 show that our method achieves the perfect detect rate of 100%, the average velocity difference of 6.9762 mph on freeways, and 8.9144 mph on both freeways and urban roads.
AB - In this paper, we propose our method for vehicle detection with multiple adaptive vehicle detectors and velocity estimation with landmark-based scanlines. Inspired by the idea for tiny object detection, we use Faster R-CNN with Resnet-101 to create different specialized vehicle detectors corresponding to different levels of details and poses. We propose a heuristic to check the fitness of a particular vehicle detector to a specific region in camera's view by the mean velocity direction and the mean object size. By this way, we can determine an adaptive set of appropriate vehicle detectors for each region in camera's view. Thus our system is expected to detect vehicles with high accuracy, both in precision and recall, even with tiny objects. We exploit the U.S. road rules for the length and distance of broken white lines on roads to propose our method for vehicle's velocity estimation using such landmarks. We determine equally-distributed scanlines, virtual parallel lines that are nearly-perpendicular to the road direction, with reference to the line connecting the corresponding ends of multiple broken white lines. From the timespan for a vehicle to cross two consecutive virtual scanlines, we can calculate the average vehicle's velocity within that road segment. We also refine the speed estimation by detecting when a vehicle stops at a traffic light, and smooth the results with a moving average filter. Experiments on the dataset of Traffic Flow Analysis from NVIDIA AI City Challenge 2018 show that our method achieves the perfect detect rate of 100%, the average velocity difference of 6.9762 mph on freeways, and 8.9144 mph on both freeways and urban roads.
UR - http://www.scopus.com/inward/record.url?scp=85060878606&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85060878606&partnerID=8YFLogxK
U2 - 10.1109/CVPRW.2018.00021
DO - 10.1109/CVPRW.2018.00021
M3 - Conference contribution
AN - SCOPUS:85060878606
T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
SP - 100
EP - 107
BT - Proceedings - 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2018
PB - IEEE Computer Society
ER -