Anomaly detection requires detecting abnormal samples in large unlabeled datasets. While progress in deep learning and the advent of foundation models has produced powerful unsupervised anomaly detection methods, their deployment in practice is often hindered by the lack of labeled data -- without it, the detection accuracy of an anomaly detector cannot be evaluated reliably. In this work, we propose a general-purpose framework for evaluating image-based anomaly detectors with synthetically generated validation data. Our method assumes access to a small support set of normal images which are processed with a pre-trained diffusion model (our proposed method requires no training or fine-tuning) to produce synthetic anomalies. When mixed with normal samples from the support set, the synthetic anomalies create detection tasks that compose a validation framework for anomaly detection evaluation and model selection. In an extensive empirical study, ranging from natural images to industrial applications, we find that our synthetic validation framework selects the same models and hyper-parameters as selection with a ground-truth validation set. In addition, we find that prompts selected by our method for CLIP-based anomaly detection outperforms all other prompt selection strategies, and leads to the overall best detection accuracy, even on the challenging MVTec-AD dataset.
Prompt learning for vision-language models, e.g., CoOp, has shown great success in adapting CLIP to different downstream tasks, making it a promising solution for federated learning due to computational reasons. Existing prompt learning techniques replace hand-crafted text prompts with learned vectors that offer improvements on seen classes, but struggle to generalize to unseen classes. Our work addresses this challenge by proposing Federated Text-driven Prompt Generation (FedTPG), which learns a unified prompt generation network across multiple remote clients in a scalable manner. The prompt generation network is conditioned on task-related text input, thus is context-aware, making it suitable to generalize for both seen and unseen classes. Our comprehensive empirical evaluations on nine diverse image classification datasets show that our method is superior to existing federated prompt learning methods, that achieve overall better generalization on both seen and unseen classes and is also generalizable to unseen datasets.
Recent advances in image captioning are mainly driven by large-scale vision-language pretraining, relying heavily on computational resources and increasingly large multimodal datasets. Instead of scaling up pretraining data, we ask whether it is possible to improve performance by improving the quality of the samples in existing datasets. We pursue this question through two approaches to data curation: one that assumes that some examples should be avoided due to mismatches between the image and caption, and one that assumes that the mismatch can be addressed by replacing the image, for which we use the state-of-the-art Stable Diffusion model. These approaches are evaluated using the BLIP model on MS COCO and Flickr30K in both finetuning and few-shot learning settings. Our simple yet effective approaches consistently outperform baselines, indicating that better image captioning models can be trained by curating existing resources. Finally, we conduct a human study to understand the errors made by the Stable Diffusion model and highlight directions for future work in text-to-image generation.
Compliant grippers, owing to adaptivity and safety, have attracted considerable attention for unstructured grasping in real applications, such as industrial or logistic scenarios. However, accurate construction of the mathematical model depicting the bidirectional relationship between shape deformation and contact force for such grippers, such as the Fin-Ray grippers, remains stagnant to date. To address this research gap, this article devises, presents, and experimentally validates a universal bidirectional force-displacement mathematical model for compliant grippers based on the co-rotational concept, which endows such grippers with an intrinsic force sensing capability and offers a better insight into the design optimization. In Part 1 of the article, we introduce the fundamental theory of the co-rotational approach, where arbitrary large deformation of beam elements can be modeled. Its intrinsic principle enables the theoretical modeling to consider various types of configurations and key design parameters with very few assumptions made. Further, a force control algorithm is proposed, providing accurate displacement estimations of the gripper under external forces with minor computational loads. The performance of the proposed method is experimentally verified through comparison with Finite Element Analysis, where the influence of four key design parameters on the gripper s performance is investigated, facilitating systematical design optimization. Part 2 of this article demonstrating the force sensing capabilities and the effects of representative co-rotational modeling parameters on model accuracy is released in Google Drive.
Compliant grasping is an essential capability for most robots in practical applications. For compliant robotic end-effectors that commonly appear in industrial or logistic scenarios, such as Fin-Ray gripper, it still remains challenging to build a bidirectional mathematical model that mutually maps the shape deformation and contact force. Part I of this article has constructed the force-displacement relationship for design optimization through the co-rotational theory with very few assumptions. In Part II, we further devise a detailed displacement-force mathematical model, enabling the compliant gripper to precisely estimate contact force sensor-free. Specifically, the proposed approach based on the co-rotational theory can calculate contact forces from deformations. The presented displacement-control algorithm elaborately investigates contact forces and provides force feedback for a force control system of a gripper, where deformation appears as displacements in contact points. Afterward, simulation experiments are conducted to evaluate the performance of the proposed model through comparisons with the finite-element analysis (FEA). Simulation results reveal that the proposed model accurately estimates contact force, with an average error of around 5% throughout all single/multiple node cases, regardless of various design parameters (Part I of this article is released in Google Drive).
Anomaly detection (AD) tries to identify data instances that deviate from the norm in a given data set. Since data distributions are subject to distribution shifts, our concept of ``normality" may also drift, raising the need for zero-shot adaptation approaches for anomaly detection. However, the fact that current zero-shot AD methods rely on foundation models that are restricted in their domain (natural language and natural images), are costly, and oftentimes proprietary, asks for alternative approaches. In this paper, we propose a simple and highly effective zero-shot AD approach compatible with a variety of established AD methods. Our solution relies on training an off-the-shelf anomaly detector (such as a deep SVDD) on a set of inter-related data distributions in combination with batch normalization. This simple recipe--batch normalization plus meta-training--is a highly effective and versatile tool. Our results demonstrate the first zero-shot anomaly detection results for tabular data and SOTA zero-shot AD results for image data from specialized domains.
Selecting informative data points for expert feedback can significantly improve the performance of anomaly detection (AD) in various contexts, such as medical diagnostics or fraud detection. In this paper, we determine a set of theoretical conditions under which anomaly scores generalize from labeled queries to unlabeled data. Motivated by these results, we propose a data labeling strategy with optimal data coverage under labeling budget constraints. In addition, we propose a new learning framework for semi-supervised AD. Extensive experiments on image, tabular, and video data sets show that our approach results in state-of-the-art semi-supervised AD performance under labeling budget constraints.
High-speed communication and accurate sensing are of vital importance for future transportation system. Integrated sensing and communication (ISAC) system has the advantages of high spectrum efficiency and low hardware cost, satisfying the requirements of sensing and communication. Therefore, ISAC is considered to be a promising technology in the future transportation system. However, due to the low transmit power of signal and the influence of harsh transmission environment on radar sensing, the signal to noise ratio (SNR) at the radar receiver is low, which affects the sensing performance. This paper introduces the intelligent reflecting surface (IRS) into ISAC system. With IRS composed of M sub-surfaces implemented on the surface of the target. The SNR at the radar receiver is 20lg(M) times larger than the scheme without IRS. Correspondingly, radar detection probability is significantly improved, and Cramer-Rao Lower Bound (CRLB) for ranging and velocity estimation is reduced. This paper proves the efficiency of IRS enabled ISAC system, which motivates the implementation of IRS to enhance the sensing capability in ISAC system.
Most vision-and-language pretraining research focuses on English tasks. However, the creation of multilingual multimodal evaluation datasets (e.g. Multi30K, xGQA, XVNLI, and MaRVL) poses a new challenge in finding high-quality training data that is both multilingual and multimodal. In this paper, we investigate whether machine translating English multimodal data can be an effective proxy for the lack of readily available multilingual data. We call this framework TD-MML: Translated Data for Multilingual Multimodal Learning, and it can be applied to any multimodal dataset and model. We apply it to both pretraining and fine-tuning data with a state-of-the-art model. In order to prevent models from learning from low-quality translated text, we propose two metrics for automatically removing such translations from the resulting datasets. In experiments on five tasks across 20 languages in the IGLUE benchmark, we show that translated data can provide a useful signal for multilingual multimodal learning, both at pretraining and fine-tuning.
We propose a parallel massage robot with compliant joints based on the series elastic actuator (SEA), offering a unified force-position control approach. First, the kinematic and static force models are established for obtaining the corresponding control variables. Then, a novel force-position control strategy is proposed to separately control the force-position along the normal direction of the surface and another two-direction displacement, without the requirement of a robotic dynamics model. To evaluate its performance, we implement a series of robotic massage experiments. The results demonstrate that the proposed massage manipulator can successfully achieve desired forces and motion patterns of massage tasks, arriving at a high-score user experience.