Seizure events can manifest as transient disruptions in the control of movements which may
be organized in distinct behavioral sequences, accompanied or not by other observable …

While head-mounted devices are becoming more compact they provide egocentric views
with significant self-occlusions of the device user. Hence existing methods often fail to …

In-context learning provides a new perspective for multi-task modeling for vision and NLP.
Under this setting the model can perceive tasks from prompts and accomplish them without …

Recent transformer-based approaches have demonstrated excellent performance in 3D
human pose estimation. However, they have a holistic view and by encoding global …

What will the future be? We wonder! In this survey, we explore the gap between current
research in egocentric vision and the ever-anticipated future, where wearable computing …

Computer vision is a powerful tool for healthcare applications since it can provide objective
diagnosis and assessment of pathologies, not depending on clinicians' skills and …

Modern multi-layer perceptron (MLP) models have shown competitive results in learning
visual representations without self-attention. However, existing MLP models are not good at …

Human pose analysis has garnered significant attention within both the research community
and practical applications, owing to its expanding array of uses, including gaming, video …

Transformers have been successfully applied in the field of video-based 3D human pose
estimation. However the high computational costs of these video pose transformers (VPTs) …

While current methods have shown promising progress on estimating 3D human motion
from monocular videos their motion estimates are often physically unrealistic because they …