HyperAI초신경
홈뉴스최신 연구 논문튜토리얼데이터셋백과사전SOTALLM 모델GPU 랭킹컨퍼런스
전체 검색
소개
HyperAI초신경

Dense Video Captioning On Activitynet

평가 지표

BLEU-3
BLEU-4
METEOR

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름
BLEU-3
BLEU-4
METEOR
Paper TitleRepository
iPerceive (Chadha et al., 2020)2.931.297.87iPerceive: Applying Common-Sense Reasoning to Multi-Modal Dense Video Captioning and Video Question Answering-
ADV-INF + Global-9.4516.36Global Object Proposals for Improving Multi-Sentence Video Descriptions
MDVC2.61.077.31Multi-modal Dense Video Captioning
PDVC (TSP features, no SCST)-2.179.03End-to-End Dense Video Captioning with Parallel Decoding
Bi-directional+intra captioning--11.28Team RUC_AIM3 Technical Report at Activitynet 2020 Task 2: Exploring Sequential Events Detection for Dense Video Captioning-
Vid2Seq--17Vid2Seq: Large-Scale Pretraining of a Visual Language Model for Dense Video Captioning
TSRM-CMG-HRNN+SCST--9.71Dense-Captioning Events in Videos: SYSU Submission to ActivityNet Challenge 2020
TSP4.162.028.75TSP: Temporally-Sensitive Pretraining of Video Encoders for Localization Tasks
GVL--10.03Learning Grounded Vision-Language Representation for Versatile Understanding in Untrimmed Videos
BMT3.841.888.44A Better Use of Audio-Visual Cues: Dense Video Captioning with Bi-modal Transformer
CM²--8.55Do You Remember? Dense Video Captioning with Cross-Modal Memory Retrieval
VTimeLLM---VTimeLLM: Empower LLM to Grasp Video Moments
0 of 12 row(s) selected.