MO1.L11.2
IMPROVING OPEN-WORLD CLASS-AGNOSTIC OBJECT DETECTORS VIA FEATURE DISTILLATION WITH STUDENT-AWARE ADAPTATION
Kengo Murata, Takuma Yamamoto, Yuya Obinata, Osafumi Nakayama, Fujitsu Limited, Japan
Session:
MO1.L11: Industry 1 Spotlight
Track:
Industry
Location:
Dena'ina: Tikahtnu Ballroom D/E
Presentation Time:
Mon, 15 Sep, 08:50 - 09:10 Anchorage Time (UTC -8)
Session Chair:
Shan Liu, Tencent Media Lab
Session MO1.L11
MO1.L11.1: DUAL-STREAM SPATIO-TEMPORAL ACCIDENT ANTICIPATION AND DETECTION
Zongyao Li, Satoshi Yamazaki, Jianquan Liu, NEC Corporation, Japan
MO1.L11.2: IMPROVING OPEN-WORLD CLASS-AGNOSTIC OBJECT DETECTORS VIA FEATURE DISTILLATION WITH STUDENT-AWARE ADAPTATION
Kengo Murata, Takuma Yamamoto, Yuya Obinata, Osafumi Nakayama, Fujitsu Limited, Japan
MO1.L11.3: Fast Bounding Box Hierarchy
Zhe Zhu, Sumit Jha, Joonsoo Kim, Arshita Gupta, Tien Bau, Samsung Research America, United States
Contacts