I'll be presenting my internship work at @Apple: "HumMUSS: Human Motion Understanding using State Space Models" next week at #CVPR2024 with Stefano @DenisTome. Looking forward to the discussions.
Poster: https://t.co/uFJ1Xrwvk8
Paper: https://t.co/iD9MVnFxDb
Comparison between HumMUSS and Motion-BERT 3D pose estimation performance (MPJPE in mm) on MPI-INF-3DHP at different sub-sampling rates. Left: causal models; Right: bi-directional models.
Introducing 'HumMUSS: Human Motion Understanding using State Space Models'! Our new model offers adaptable frame rates, faster real-time predictions, and high accuracy in motion analysis. https://t.co/BdOobiriZP
#ComputerVision#StateSpaceModels#CVPR2024
HumMUSS efficiently utilizes only the current frame and its current state for predictions, ensuring constant memory and time complexity. HumMUSS also generalizes to new frame rates and enhances the training speed without compromising the prediction accuracy.
Excited to share that our demo won the Best Demo Award at @CVPR 2022!
Speech Driven Tongue Animation, "improve realistic animations & surpass the uncanny valley by enabling data-driven tongue animation"
@Grekogecko@carsten_epic@iainmatthews
Project: https://t.co/DOy07GJgqY
Flying to New Orleans for #CVPR2022
Ping me if you are interested in having a chat/drinking some coffee!
Looking forward to meeting a lot of old friends.