【公布得獎名單】Embedded Deep Learning Object Detection Model Competition in IEEE MMSP 2019
2019-08-12
PAIR Labs與交大電子系郭峻因教授團隊合作舉辦之競賽Embedded Deep Learning Object Detection Model Competition in IEEE MMSP 2019,比賽最終結果已出爐,由中國湖南湘潭大學的團隊(R.JD)獲得冠軍!
本次比賽由郭峻因教授團隊提供自建之道路圖像資料作為比賽訓練資料集,內容涵蓋行人、車輛及騎士,並邀請選手據此設計適合在嵌入式系統進行物件偵測的模型。參賽隊伍來自台灣、新加坡、中國等地,超過80個團隊報名參賽,總共有22支隊伍進入初賽,9支隊伍進入決賽。最終得獎名單公告如下:
-
- 冠軍: R.JD
- 亞軍: 從缺 (成績未超過0.5)
- 季軍: 從缺 (成績未超過0.5)
另外,決賽成績達0.46以上的隊伍將受邀在MMSP 2019研討會中發表本次比賽相關論文,邀請隊伍名單如下:
-
- R.JD
- nctuai
- chenjiaqi
- IMMVP
◆ 詳細決賽成績請參閱本表(Final Evaluation Result):
Name | Affiliation | mAP | Model Size
(MByte) |
Complexity
(GOPS/frame) |
Speed
(ms/frame) |
Award | MMSP paper invitation |
IMMVP | Institute of Information Science, Academia Sinica | 0.4605 | 57.1 | 724.9 | 510.3 | Not qualified (mAP<0.50) | YES |
NPUST-MIS-No.1 | Department of Management Information Systems, National Pingtung University of Science and Technology | 0.4396 | 238.4 | 115.6 | 514.5 | Not qualified (mAP<0.50) | —– |
R.JD | The College of Information Engineering of Xiangtan University | 0.5389 | 124.6 | 43.6 | 460.5 | Champion | YES |
nctuai | Department of Electrical Engineering, National University of Tainan | 0.4760 | 195.2 | 490.1 | 1338.9 | Not qualified (mAP<0.50) | YES |
chenjiaqi | The College of Information Engineering of Xiangtan University | 0.4619 | 114.0 | 339.5 | 1195.0 | Not qualified (mAP<0.50) | YES |