Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper addresses the challenge of standardizing performance across GelSight Mini sensors by introducing a method that enables zero-shot transfer of 3D force map estimation. The approach involves a two-stage process: first, reconstructing tactile images into a generalized format using a UniT-based model, followed by 3D force map estimation with a U-Net network. The method demonstrates high fidelity in image reconstruction (SSIM of 0.9338) and accurate force estimation (MAE_F of 1.1294 N), showcasing its effectiveness across different sensor units.
Zero-shot transfer of force map estimation could eliminate the need for retraining models for each GelSight Mini sensor, streamlining production and deployment.
Despite the rapid industrialization of the touch sensor manufacturing process, most of these sensors are still handmade in research laboratories. This complicates standardizing their performance, requiring the repetition of data collection and training models for each unit produced. To address this problem, this paper presents a method that can generalize the estimation of 3D force maps across different GelSight Mini sensor units, regardless of the sensor version. Specifically, the method consists of two stages: a domain adaptation stage, in which the input tactile image is reconstructed as a general tactile image using a UniT-based model; and a stage for estimating 3D force maps employing a U-Net network. Our proposal achieves promising results in both steps, such as an SSIM of 0.9338 +- 0.0358 in the image reconstruction phase and an MAE_F of 1.1294 +- 1.5934(N) in the force estimation phase.