AI Paper Atlas

Research map / Benchmarks / ActivityNet-Captions

ActivityNet-Captions leaderboard

34 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ActivityNet-Captions. Best mIoU: 56.25 (EMB (ELA, multi-candidate), ECCV 2022).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

ActivityNet-Captions · mIoU

17 papers, each with its best reported number for its own method; higher is better.

#mIoU ↑MethodPaperTopic
156.25EMB (ELA, multi-candidate)
VSLNet
Video Activity Localisation with Uncertainties in Temporal Boundary
ECCV 2022 · Table 2
Video-language understanding
245.86Three-stream DRFT
LGI
End-to-end Multi-modal Video Temporal Grounding
NeurIPS 2021 · Table 2
Video-language understanding
345.7+F2G-FT
Qwen3-VL-8B
Foresee-to-Ground: From Predictive Temporal Perception to Evidence-Driven Reasoning for Video Temporal Grounding
ICML 2026 · Table 1
Video-language understanding
442.11BPNet
VSLNet backbone
Boundary Proposal Network for Two-stage Natural Language Video Localization
AAAI 2021 · Table: ActivityNet Captions re
Video-language understanding
541.13LGI (Ours)Local-Global Video-Text Interactions for Temporal Grounding
CVPR 2020 · Table 2
Video-language understanding
640.86HUAL (50%, 5)
SeqPAN
Are Binary Annotations Sufficient? Video Moment Retrieval via Hierarchical Uncertainty-Based Active Learning
CVPR 2023 · Table 1
Video-language understanding
740.65CPNetProposal-Free Video Grounding with Contextual Pyramid Network
AAAI 2021 · Table 3
Video-language understanding
839.8GDPRethinking the Bottom-Up Framework for Query-Based Video Localization
AAAI 2020 · Table 1
Video-language understanding
936.99ABLR full-awTo Find Where You Talk: Temporal Sentence Localization in Video with Attention Based Location Regression
AAAI 2019 · Table: Comparison of different
Video-language understanding
1036.85CBPTemporally Grounding Language Queries in Videos by Contextual Boundary-Aware Prediction
AAAI 2020 · Table: Performance comparison
Video-language understanding
7 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

ActivityNet-Captions · R@1 IoU=0.5

17 papers, each with its best reported number for its own method; higher is better.

7 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

ActivityNet-Captions · R@0.5

10 papers, each with its best reported number for its own method; higher is better.

#R@0.5 ↑MethodPaperTopic
146.1+F2G-FT
Qwen3-VL-8B
Foresee-to-Ground: From Predictive Temporal Perception to Evidence-Driven Reasoning for Video Temporal Grounding
ICML 2026 · Table 1
Video-language understanding
245.99Ours (w/ Swin-B)
Swin-B
ProTeGe: Untrimmed Pretraining for Video Temporal Grounding by Video Temporal Grounding
CVPR 2023 · Table 1b
Video-language understanding
345.72Three-stream DRFT
LGI
End-to-end Multi-modal Video Temporal Grounding
NeurIPS 2021 · Table 2
Video-language understanding
441.51LGI (Ours)Local-Global Video-Text Interactions for Temporal Grounding
CVPR 2020 · Table 2
Video-language understanding
538.09HUAL (50%, 5)
SeqPAN
Are Binary Annotations Sufficient? Video Moment Retrieval via Hierarchical Uncertainty-Based Active Learning
CVPR 2023 · Table 1
Video-language understanding
632.98MMDist TeacherLocal-Global Multi-Modal Distillation for Weakly-Supervised Temporal Video Grounding
AAAI 2024 · Table 1
Video-language understanding
730.56BTDP
EMB
Boundary-Aware Temporal Dynamic Pseudo-Supervision Pairs Generation for Zero-Shot Natural Language Video Localization
AAAI 2025 · Table 1
Video-language understanding
830.08PSVL (PQ+TEP)
zero-shot, no supervision
Zero-Shot Natural Language Video Localization
ICCV 2021 · Table 2
Video-language understanding
928.19CORONETCommonsense for Zero-Shot Natural Language Video Localization
AAAI 2024 · Table 1
Video-language understanding
1023Momentor (7B)
LLaMA 7B
Momentor: Advancing Video Large Language Model with Fine-Grained Temporal Reasoning
ICML 2024 · Table: Comparison with existin
Video-language understanding

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.