Prajwal Gatti
@prajwalgatti.bsky.social
CS PhD student at the University of Bristol.
Thrilled to announce HD-EPIC 🎉 a highly-detailed egocentric validation dataset with wide range of manual annotations and a new VQA benchmark that challenges the latest VLMs. Pre-print and dataset now available. Check it out 🧑🍳
🛑📢
HD-EPIC: A Highly-Detailed Egocentric Video Dataset
hd-epic.github.io
arxiv.org/abs/2502.04144
New collected videos
263 annotations/min: recipe, nutrition, actions, sounds, 3D object movement &fixture associations, masks.
26K VQA benchmark to challenge current VLMs
1/N
HD-EPIC: A Highly-Detailed Egocentric Video Dataset
hd-epic.github.io
arxiv.org/abs/2502.04144
New collected videos
263 annotations/min: recipe, nutrition, actions, sounds, 3D object movement &fixture associations, masks.
26K VQA benchmark to challenge current VLMs
1/N
February 7, 2025 at 12:38 PM
Thrilled to announce HD-EPIC 🎉 a highly-detailed egocentric validation dataset with wide range of manual annotations and a new VQA benchmark that challenges the latest VLMs. Pre-print and dataset now available. Check it out 🧑🍳
Reposted by Prajwal Gatti
Last working day in 2024 @bristoluni.bsky.social wishing you all happy holidays from myself @michaelwray.bsky.social @prajwalgatti.bsky.social M Wray, T Perrett, Z Zhu, A Fragomeni, S Pollard, K Flanagan, S Bansal, S Sinha, M Tatano, M Hateno, R Guerrier, A Darkhalil, K Parida, J Chalk, F Abdelazim
December 20, 2024 at 12:35 PM
Last working day in 2024 @bristoluni.bsky.social wishing you all happy holidays from myself @michaelwray.bsky.social @prajwalgatti.bsky.social M Wray, T Perrett, Z Zhu, A Fragomeni, S Pollard, K Flanagan, S Bansal, S Sinha, M Tatano, M Hateno, R Guerrier, A Darkhalil, K Parida, J Chalk, F Abdelazim
Now on ArXiv
ShowHowTo: Generating Scene-Conditioned Step-by-Step Visual Instructions
arxiv.org/abs/2412.01987
soczech.github.io/showhowto/
Given one real image &variable sequence of text instructions, ShowHowTo generates a multi-step sequence of images *conditioned on the scene in the REAL image*
🧵
ShowHowTo: Generating Scene-Conditioned Step-by-Step Visual Instructions
arxiv.org/abs/2412.01987
soczech.github.io/showhowto/
Given one real image &variable sequence of text instructions, ShowHowTo generates a multi-step sequence of images *conditioned on the scene in the REAL image*
🧵
December 5, 2024 at 3:11 PM