5 accepted papers
Sign Language Translation (SLT) converts continuous sign videos into spoken language text, yet current models, whether gloss-based or gloss-free, struggle with long or discourse-level inputs. Recent architectures such as TwoStreamNetwork and CV-SLT have nearly saturated short-sentence accuracy, but
Deep learning (DL) models are increasingly deployed in safety-critical applications such as face recognition, autonomous driving, and medical diagnosis. Despite their impressive accuracy, they remain vulnerable to adversarial examples - subtle perturbations that can cause incorrect predictions, i.e.
There are many instances of helical mechanisms that are used to efficiently grasp different objects with various shapes and sizes in nature. Inspired by the helical grasping in the nature, we proposed a helical bistable soft gripper with high load capacity and energy saving. An off-the-shelf bistabl…
Legged robots are capable of achieving energy-efficiency by introducing elastic elements in parallel with the actuators during the stance phase. However, the parallel elastic actuation results in a coupling motion between the parallel spring and the flexion movement of the leg mechanism during the s
Magnetorheological (MR) fluids have applications in diverse areas such as automobile suspension, architecture, and vibration control due to their fast response, wide range of controllability, and temperature resistance. However, traditional MR fluids-based applications for robotics lack the capabili