TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Datasets/EPIC-KITCHENS-100

EPIC-KITCHENS-100

TextsVideosCC BY NC 4.0Introduced 2020-06-23

This paper introduces the pipeline to scale the largest dataset in egocentric vision EPIC-KITCHENS. The effort culminates in EPIC-KITCHENS-100, a collection of 100 hours, 20M frames, 90K actions in 700 variable-length videos, capturing long-term unscripted activities in 45 environments, using head-mounted cameras. Compared to its previous version (EPIC-KITCHENS-55), EPIC-KITCHENS-100 has been annotated using a novel pipeline that allows denser (54% more actions per minute) and more complete annotations of fine-grained actions (+128% more action segments). This collection also enables evaluating the "test of time" - i.e. whether models trained on data collected in 2018 can generalise to new footage collected under the same hypotheses albeit "two years on". The dataset is aligned with 6 challenges: action recognition (full and weak supervision), action detection, action anticipation, cross-modal retrieval (from captions), as well as unsupervised domain adaptation for action recognition. For each challenge, we define the task, provide baselines and evaluation metrics.

Benchmarks

2D Human Pose Estimation/Recall@52D Human Pose Estimation/Top-5 Verb2D Human Pose Estimation/Top-5 NounAction Anticipation/Recall@5Action Anticipation/Top-5 VerbAction Anticipation/Top-5 NounAction Localization/Avg mAP (0.1-0.5)Action Localization/mAP IOU@0.1Action Localization/mAP IOU@0.2Action Localization/mAP IOU@0.3Action Localization/mAP IOU@0.4Action Localization/mAP IOU@0.5Action Recognition/Action@1Action Recognition/Verb@1Action Recognition/Noun@1Action Recognition/GFLOPsAction Recognition/Recall@5Action Recognition/Top-5 VerbAction Recognition/Top-5 NounAction Recognition/HMAction Recognition In Videos/Recall@5Action Recognition In Videos/Top-5 VerbAction Recognition In Videos/Top-5 NounActivity Recognition/Action@1Activity Recognition/Verb@1Activity Recognition/Noun@1Activity Recognition/GFLOPsActivity Recognition/Recall@5Activity Recognition/Top-5 VerbActivity Recognition/Top-5 NounActivity Recognition/HMAudio Classification/Top-1 ActionAudio Classification/Top-1 NounAudio Classification/Top-1 VerbAudio Classification/Top-5 ActionAudio Classification/Top-5 NounAudio Classification/Top-5 VerbClassification/Top-1 ActionClassification/Top-1 NounClassification/Top-1 VerbClassification/Top-5 ActionClassification/Top-5 NounClassification/Top-5 VerbDomain Adaptation/Average AccuracyTemporal Action Localization/Avg mAP (0.1-0.5)Temporal Action Localization/mAP IOU@0.1Temporal Action Localization/mAP IOU@0.2Temporal Action Localization/mAP IOU@0.3Temporal Action Localization/mAP IOU@0.4Temporal Action Localization/mAP IOU@0.5Unsupervised Domain Adaptation/Average AccuracyVideo/Avg mAP (0.1-0.5)Video/mAP IOU@0.1Video/mAP IOU@0.2Video/mAP IOU@0.3Video/mAP IOU@0.4Video/mAP IOU@0.5Zero-Shot Learning/Avg mAP (0.1-0.5)Zero-Shot Learning/mAP IOU@0.1Zero-Shot Learning/mAP IOU@0.2Zero-Shot Learning/mAP IOU@0.3Zero-Shot Learning/mAP IOU@0.4Zero-Shot Learning/mAP IOU@0.5

Related Benchmarks

EPIC-KITCHENS-100 (test)/2D Human Pose Estimation/recall@5EPIC-KITCHENS-100 (test)/Action Anticipation/recall@5EPIC-KITCHENS-100 (test)/Action Recognition/recall@5EPIC-KITCHENS-100 (test)/Action Recognition In Videos/recall@5EPIC-KITCHENS-100 (test)/Activity Recognition/recall@5

Statistics

Papers
162
Benchmarks
63

Links

Homepage

Tasks

2D Human Pose EstimationAction AnticipationAction LocalizationAction RecognitionAction Recognition In VideosActivity RecognitionAudio ClassificationClassificationDomain AdaptationMulti-Instance RetrievalOpen Vocabulary Action RecognitionTemporal Action LocalizationUnsupervised Domain AdaptationVideoZero-Shot Learning