PressureVision: Estimating Hand Pressure from a Single RGB Image

Abstract

PressureVisionNet takes a single RGB image as input and outputs a pressure image that is an estimate of the pressure applied by the hand to a flat contact surface.

Visual Pressure Estimation and Control for Soft Robotic Grippers

“…Our work builds on prior efforts to visually infer pressure applied by human hands [1]. We use the same neural network architecture, but apply it to inference and control of soft robotic grippers.…”

“…To evaluate the performance of VPEC-Net, we perform evaluations on the held-out test set. We use a variety of evaluation metrics similar to [1] to quantify pressure estimation accuracy.…”

“…Overall, our results show that VPEC enables grippers with high compliance to perform precision manipulation. 1…”