Papers with Code · CC BY-SA 4.0

AVA-ActiveSpeaker

아카이브 평가 20건 · 마지막 적재/변경 2026-10-30

평가 기록

과거 수집 기록이며 최신 순위가 아닙니다. 데이터셋 버전·split·실험 조건과 지표 방향이 검증되지 않은 결과는 직접 비교할 수 없습니다. 날짜는 논문 날짜이며 최신 평가일을 뜻하지 않습니다.

표가 화면보다 넓으면 좌우로 스크롤하여 논문과 구현 링크를 확인하세요.

원본 기록 순서 · 순위 미부여
모델	원본 지표·값	논문	구현
LoCoNet+TalkNCE	validation mean average precision 95.5%	TalkNCE: Improving Active Speaker Detection with Talk-Aware Contrastive Learning 2023-09-21	코드 1
LoCoNet + Laser	validation mean average precision 95.3%	LASER: Lip Landmark Assisted Speaker Detection for Robustness 2025-01-21	코드 1
LoCoNet	validation mean average precision 95.2%	LoCoNet: Long-Short Context Network for Active Speaker Detection 2023-01-19	코드 1 · 코드 2
SPELL+	validation mean average precision 94.9%	Learning Long-Term Spatial-Temporal Graphs for Active Speaker Detection 2022-07-15	코드 1 · 코드 2
UniCon+	validation mean average precision 94.5%	UniCon+: ICTCAS-UCAS Submission to the AVA-ActiveSpeaker Task at ActivityNet Challenge 2022 2022-06-22	연결 없음
SPELL	validation mean average precision 94.2%	Learning Long-Term Spatial-Temporal Graphs for Active Speaker Detection 2022-07-15	코드 1 · 코드 2
EASEE-50	validation mean average precision 94.1%	End-to-End Active Speaker Detection 2022-03-27	코드 1 · 코드 2 · 코드 3
Light-ASD	validation mean average precision 94.1%	A Light Weight Model for Active Speaker Detection 2023-03-08	코드 1
Extended UniCon	validation mean average precision 93.6%	ICTCAS-UCAS-TAL Submission to the AVA-ActiveSpeaker Task at ActivityNet Challenge 2021 2021-06-01	연결 없음
ASDNet	validation mean average precision 93.5%	How to Design a Three-Stage Architecture for Audio-Visual Active Speaker Detection in the Wild 2021-06-07	코드 1
GSCMIA	validation mean average precision 92.86%	Audio-Visual Activity Guided Cross-Modal Identity Association for Active Speaker Detection 2022-12-01	코드 1
TalkNet	validation mean average precision 92.3%	NUS-HLT Report for ActivityNet Challenge 2021 AVA (Speaker) 2021-06-01	코드 1
UniCon	validation mean average precision 92.0%	UniCon: Unified Context Network for Robust Active Speaker Detection 2021-08-05	연결 없음
SA-uncertainty Fusion	validation mean average precision 91.9%	Active Speaker Detection as a Multi-Objective Optimization with Uncertainty-based Multimodal Fusion 2021-06-07	연결 없음
VTP (visual only)	validation mean average precision 89.2%	Sub-word Level Lip Reading With Visual Attention 2021-10-14	연결 없음
MAAS-TAN	validation mean average precision 88.8%	MAAS: Multi-modal Assignation for Active Speaker Detection 2021-01-11	코드 1
VGG-{LSTM+TCN} (ensemble)	validation mean average precision 87.8%	Naver at ActivityNet Challenge 2019 -- Task B Active Speaker Detection (AVA) 2019-06-25	연결 없음
Active Speakers in Context	validation mean average precision 87.1%	Active Speakers in Context 2020-05-20	코드 1
MAAS-LAN	validation mean average precision 85.1%	MAAS: Multi-modal Assignation for Active Speaker Detection 2021-01-11	코드 1
3D-ResNet-GRU	validation mean average precision 84.0%	Multi-Task Learning for Audio Visual Active Speaker Detection 2019-06-01	연결 없음