This study introduces a new force-position impedance control strategy tailored for robotic systems engaged in the manipulation of elastic garments. Our approach is innovative for incorporating external force and impedance control within an incremental trajectory generation framework. This method allows for dynamic adjustment to prevent over-deformation and potential damage during garment handling, enabling the fabric to deform to desired states and follow intended trajectory trends without compromising textile integrity. By utilizing external forces in an adaptive, incremental manner, we offer a nuanced solution for expected garment manipulation. Our strategy demonstrates the feasibility of detailed manipulation tasks in industrial settings, bridging the gap between simulated environments and real-world operations. This effort underscores the importance of automation in enhancing textile handling processes, focusing on practical applications rather than theoretical advancements.

This research unveils a novel Multi-Critic Reinforcement Learning framework designed to navigate the multifaceted challenges associated with multi-phased garment handling tasks, notably marked by persistent contact and erratic deformations between textiles and solid bodies. These tasks, ubiquitous in domestic and industrial environments, encompass activities such as dressing, fabric printing, and pressing, and are complicated by the unpredictability of textile states and the intricacy of devising control strategies. Our reinforcement learning model combines multiple time-sequenced Critic networks with traditional Deep Deterministic Policy Gradient (DDPG) techniques, thereby equipping the system to adapt to the diverse effects of fabric distortions throughout various stages. The effectiveness of this approach is demonstrated through a multi-phase pre-printing operation and further validated by real-world implementations, showing significant improvements in coverage and a substantial reduction in wrinkle formation, with its versatility further confirmed by a complex vertical dressing task. We anticipate future applications of this framework in a range of complex problems, not just garment handling. The model used in this paper can be found at https://github.com/jkk5454/multiddpg.git.

Cross-spectral face hallucination is an intuitive way to mitigate the modality discrepancy in Heterogeneous Face Recognition (HFR). However, due to imaging differences, the hallucination inevitably suffers from a shape misalignment between paired heterogeneous images. Rather than building complicated architectures to circumvent the problem like previous works, we propose a simple yet effective method called Shape Alignment FacE (SAFE). Specifically, given an image, we align its shape to that of the paired one under the assistance of a 3D face model. The produced aligned pair enables us to train a lightweight generator that solely concentrates on spectrum translation with a pixel-wise supervision. However, since the 3D face model is powerless to attributes like the hair and glasses, there are still pixel discrepancies between the aligned pair. Given that, in the image space, we introduce a probabilistic pixel-wise loss that incorporates the discrepancies into a probabilistic distribution. Moreover, in order to alleviate the influence of the shape misalignment on spectrum translation, a spectrum optimal transport is performed in a shape-irrelevant latent space. Note that, in the final inference phase, except the lightweight generator, all other auxiliary modules are discarded. In addition to superior performance in qualitative synthesis and quantitative recognition, extensive experiments on 6 datasets demonstrate that our method also gains other two distinct advantages over existing state-of-the-art counterparts. The first is using a more lightweight generator. Compared with the state-of-the-art method, our method can achieve higher recognition results with 128x fewer parameters and 63x fewer FLOPs with only 4.58 ms latency on a single TITAN-XP. The second is training on low-shot datasets such as Oulu-CASIA NIR-VIS that just contains 1,920 images from 20 identities. To the best of our knowledge, we are the first that can perform well on such a small-scale dataset. These advantages make our method more practical in the real world and further push boundaries of heterogeneous face recognition.

Heterogeneous face recognition (HFR) aims at matching face images across different domains. It is challenging due to the severe domain discrepancies and overfitting caused by small training datasets. Some researchers apply a “recognition via generation” strategy and propose to solve the problem by translating images from a given domain into the visual domain. However, in many HFR tasks such as near-infrablack HFR, there is no paiblack data, which makes it an unsupervised generation. Pose variations, background differences, and many other factors present challenges. Moreover, the generated results lack diversity since many previous works regard this image translation as a “one-to-one” generation task. Considering the information deficiency in the input images, we propose to formulate this image translation process as a “one-to-many” generation problem. Specifically, we introduce reference images to guide the generation process. We propose a memory module to explore the prototypical style patterns of the reference domain. After self-supervised updating, the memory items are attentively aggregated to represent the style information. Moreover, to subtly fuse the contents of input images with the style of reference images, we propose a novel style transformer module. Specifically, we crop the encoded input and reference feature maps into patches, and use the style transformer to establish long-range dependencies between the input and reference patches. Thus, the style of every input patch is transferblack based on those of the most relevant reference patches. Extensive experiments on multiple datasets for various HFR tasks, including NIR-VIS, thermal-VIS, sketch-photo, and gray-RGB, are conducted. The robustness and effectiveness of the proposed MMTN are demonstrated both quantitatively and qualitatively.