Last 24 Hours
5 TaskRound
13 TaskLogs
6352 MB
Last 7 Days
9 TaskRound
26 TaskLogs
6358 MB
All Time
3 Tasks
363 TaskRounds
12204 TaskLogs
134959 MB
Aerial Traffic Patrol
June 23, 2025
The team from Purdue University have developed δ-ATP with advanced computer vision techniques customized for aerial traffic patrol so as to improve vehicle detection/tracking and movement speed monitoring in a faster and more reliable manner.
- Dataset (public release at GitHub) (March 2025, West Lafayette, IN) This dataset is collected via AirLab, including (1) high-resolution aerial videos (30 FPS, 4K) and (2)ground truth speed captured by radar gun. Dataset also available at OneDrive.
- Jiaxin Du, Chunyi Peng, δ-ATP: Towards Fast and Reliable Aerial Traffic Patrol, DroNet'25, June 27, 2025, Anaheim, CA, USA.
- Jiaxin Du, Shilong Lei, Chunyi Peng, Poster: D-AirPatrol: A Dual-Layer Architecture for Traffic Patrol From the Sky, Proceedings of the 30th Annual International Conference on Mobile Computing and Networking (ACM MobiCom'24), November 18-22, 2024, Washington D.C., USA (ACM MobiCom'24 Best Poster Award)
Autonomous Drone Delivery to Door
July 17, 2025
The team from Purdue University have developed two novel patches to the existing semantic segmentation technique called Structural Semantic Segmentation (SSS) and Robust Structural Semantic Segmentation (RSSS) to assist the last-hundred-feet problem in autonomous drone delivery to door.
- Demo (HotMobile'24 Best Demo Award, Feb 2024). More at Github.
- Shengqing Xia, Jiaxin Du, Chunyi Peng, RSSS: Robust Structural Semantic Segmentation for Autonomous Drone Delivery to Door, IROS'25, Hangzhou, China.
- Shengqing Xia, Junpeng Guo, Chunyi Peng, SSS: Towards Autonomous Drone Delivery to Your Door Over House-Aware Semantics, HotMobile'24, San Diego, CA, USA (with ACM HotMobile'24 Best Demo Award)