datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
wds-mops-kitchen-100kThis is the anonymous Dataset upload for Multi-Objective Photoreal Simulation (MOPS) Dataset [under review].
Official code available at: https://github.com/mopsneurips26-code/mops-submission
epic_kitchen_videomae_L14_feature_fps8
📙 Overview
EPIC-Kitchen-100 video features extracted by VideoMAE_L14 at 8 fps. It is used for evaluating the video-text retrieval ability of EgoInstructor.
It contains 700 files, each file (e.g. P01_01.pth.tar) is a TxD feature vector, where T refers to the length of the video and D is 768.
🏋️ How-To-Use
Please refer to code EgoInstructor for details.
🎓 Citation
@article{xu2024retrieval,
title={Retrieval-augmented egocentric video captioning},
author={Xu… See the full description on the dataset page: https://huggingface.co/datasets/Jazzcharles/epic_kitchen_videomae_L14_feature_fps8.octo_ucsd_kitchenepic_kitchen_100_resume
EK100 V-JEPA2 Cached Tokens
Cached official V-JEPA2 ViT-g/384 action-anticipation tokens for EPIC-Kitchens-100.
Each batch_*.pt payload contains:
tokens: [B, 9792, 1408], fp16, context encoder tokens plus predicted future tokens.
verb_labels, noun_labels, action_labels: canonical labels used by the cached probe pipeline.
anticipation_time: seconds into the future, currently 1.0.
metadata: video id, narration id, frame anchors, and source video path.
Generation settings:
frames… See the full description on the dataset page: https://huggingface.co/datasets/sjmathy/epic_kitchen_100_resume.