Smart manufacturing based on artificial intelligence and information communication technology will become the main contributor to the digital economy of the upcoming decades. In order to execute flexible production, smart manufacturing must holistically integrate wireless networking, computing, and automatic control technologies. This article discusses the challenges of this complex system engineering from a wireless networking perspective. Starting from enabling flexible reconfiguration of a smart factory, we discuss existing wireless technology and the trends of wireless networking evolution to facilitate multirobot smart factories. Furthermore, the special sequential decision-making of a multirobot manufacturing system is examined. Social learning can be used to extend the resilience of precision operation in a multirobot system by taking network topology into consideration, which also introduces a new vision for the cybersecurity of smart factories. A summary of highlights of technological opportunities for holistic facilitation of wireless networked multirobot smart factories rounds off this article.

The operation of autonomous multi-agent systems (MAS) under the complex interactions among intelligent agents heavily relies on the consensus protocols to facilitate multi-agent collaboration. In this paper, we novelly study the coordination and control strategy for robust consensus building in a distributed MAS of bayesian social learning agents that make sequential decisions with finite horizon. Unlike the conventional consensus problems in literature, the idiosyncratic information cascading behavior arising from multi-agent social learning poses a new challenge to consensus protocol design. Although cascading can alleviate randomness and enhance consensus, cascading in the early stages of sequential decision making may cause error-prone bifurcation behavior. Therefore, how to properly utilize the advantage of information cascade and meanwhile mitigating its undesirable side effect is critical in the design of consensus protocol to control distributed MAS of social learning agents. Considering communication topology, we uniquely develop a robust cascading strategy for agents adapting to local network neighborhood. Our results show that the proposed strategy not only mitigates early cascading failures but also guarantees multiagent consensus in various complex network topologies.

Information cascade characterizes the situation where rational agents are overloaded with social observations and can make decisions ignoring their own personal evidence. If the social observations are misleading, the overall system performance can be subsequently compromised. Therefore, it is of interests to design a socially or interactively networked system falling into undesirable or desirable cascades. In this paper, we study the driving factors of steering information cascades in a socially networked system by considering a sequential Bayesian social learning model with arbitrary network topology and utility function. In particular, we analyze the impacts from different random network topologies on the overall system performance such as error probability, fraction of cascades, and total utility. Furthermore, we demonstrate two systematic approaches, selective rewiring and incentive seeding, which can change the information structure observed by agents and hence steer information cascades. Our results show that when agents restructure their observation network and provide proper incentives for cascades in the correct decisions, the overall social system performance can be well-steered.

This paper deals with the problem of clothing retrieval in a recommendation system. We develop a hierarchical deep search framework to tackle this problem. We use a pre-trained network model that has learned rich mid-level visual representations in module 1. Then, in module 2, we add a latent layer to the network and have neurons in this layer to learn hashes-like representations while fine-tuning it on the clothing dataset. Finally, module 3 achieves fast clothing retrieval using the learned hash codes and representations via a coarse-to-fine strategy. We use a large clothing dataset where 161,234 clothes images are collected and labeled. Experiments demonstrate the potential of our proposed framework for clothing retrieval in a large corpus.

With the rapid development of e-commerce and explosive growth of online shopping market, the problem of how to optimize the process of guiding the user to the huge amount of online products have been urgent. Existing recommender systems use information from users' profiles (demographic filtering), similar neighbors (collaborative filtering), and textual description (content-based model) to make recommendations, which easily generate irrelevant suggestions to users due to the ignorance of users' intentions and the visual similarity among products. In this paper, we proposed an interactive product recommendation method, which considers not only the product diversity but also the visual similarity, to interactively capture a user's real intention and refine the product recommendation result based on the user's real product interests. Our algorithm is experimentally evaluated under a real-world user log, and shown to significantly improve recommendation accuracy over the traditional approaches.

With the number of people considered to be obese and having chronic disease rising across the world, the role of IT solution in nutrition management and planning has been receiving increased attention by medical professionals in recent years. A key factor toward a successful personalized nutrition planning is an individual's food preference instead of dogmatic nutrition pattern since it is unlikely that an individual would accept the meal plan merely based on the nutrition supplements. However, personal preferences about foods are obviously relatively harder to acquire comparing to common nutrition requirements that can be easily obtained from guidelines. In this paper, we proposed a personalized nutrition planning method based on random walks theory to maximize the planning satisfaction. The personal guidelines generated by the proposed method are expected to potentially improve the healthy diet compliance and thus reduce the risk of chronic diseases of individuals.

One current direction to enhance the search accuracy in visual object retrieval is to reformulate the original query through (pseudo-)relevance feedback, which augments a query with visual terms from the image documents most highly ranked by an initial search or identified by user. However, query and feedback images usually contain multiple objects or aspects, and as a consequence the original query's focus may drift because of the newly added terms and noises. The results of using an augmented query are thus often inferior to that of using only the original one. In this paper we propose the topic-sensitive image retrieval with noise-proof relevance feedback to address the query drift problem in visual object retrieval. The proposed method removes irrelevant noises and topics from both query and feedback images to prevent query drift. A discriminative learning strategy is then employed to re-rank and improve the initial search result. Experiments on a real world data set demonstrate the effectiveness of our approach and show that the proposed approach can better learn user intention.

Effective personal dietary guidelines are essential for health management and preventing chronic diseases. The objective of this research is to achieve nutrient-balanced food recommendations for each individual, while considering individual's requirements at the same time. To reach this goal, we developed a location-aware interactive diet consultant named SmartDiet based on the multi-objective optimization. The proposed personalized diet planning approach not only translates nutrient recommendations into realistic dish choices, but also accepts feedbacks from users to fine-tune their meal plans. The results showed that daily nutrition needs can be fulfilled by the designated meals, and the interactive diet planning scheme helps a user adjust the plan in an easier way. The guidelines generated by SmartDiet are expected to potentially improve the overall health and reduce the risk of chronic diseases of individuals.

We study the problem of event detection from realistic videos with repetitive sequential human activities. Despite the large body of work on event detection and recognition, very few have addressed low-quality videos captured from realistic environments. Our framework is based on solving the shortest path on a temporal-event graph constructed from the video content. Graph vertices correspond to detected event primitives, and edge weights are set according to generic knowledge of the event patterns and the discrepancy between event primitives based on a greedy matching of their visual features. Experimental results on videos collected from a retail environment validate the usefulness of the proposed approach.

We consider the problem of ranking refinement for image object retrieval, whose goal is to improve an existing ranking function by a small number of labeled instances. To retrieve the relevant image object, one state-of-the-art approach is to use the relevance feedback: it first ranks the images in database based on a given ranking function (i.e., base ranker), and then rerank the initial result by further introducing user's feedback information. The key challenge of combining the information from the base ranker and user's feedback comes from the fact that the base ranker tends to give an imperfect result and the information obtained from user's feedback tends to be very noisy. This paper describes an Intention-Focused Active Reranking, an approach for automatically finding the right information to re-estimate the query model. Three novel strategies are proposed to boost the performance of the base ranker: (1) an active selection criterion, which obtains a small number of feedback images that are the most informative to the base ranker for user labeling; (2) the user intention verification, which captures the user's intention in object level to alleviate the query drift problem; (3) a discriminative query model re-estimation, which augments the generative approach with a model of the discriminative information conveyed by positive and negative feedback information. Experiments on a real world data set demonstrate the effectiveness of the proposed approach and furthermore it significantly outperforms the baseline visual bag-of-words retrieval.

A new challenge in the spam email detection is the emergence of image spam, which consists in embedding the advertising messages into attached images to defeat the conventional text-based anti-spam technologies. New techniques are needed to filter these spam messages. In this paper, we proposed a prototype system to automatically classify an image directly as being spam or ham. The proposed method extracts latent topics in image to train a binary classifier for detecting spam images, and achieves more promising detection accuracy than conventional antispam approaches. In addition, a detection cascade is proposed to further reduce the computation overhead of the spam filter. Our algorithm is experimentally evaluated under a public spam image dataset, and shown to significantly improve both the detection accuracy and execution efficiency over the baseline approach.

One leading framework for image object mining is the bag-of-words (BOW) approach. The idea is to encode an image as a collection of visual words of the quantized local patches. Objects in the image can then be retrieved through inferring the semantic topics associated with the set of visual words. However, the visual BOW mining framework is apt to suffer from the so-called term-mismatch problem (a.k.a. vocabulary problem). This is caused by the poverty of query information, and consequently becomes an obstacle to deal with synonymy (i.e., different visual words for describing the same object). In this paper, we propose a novel language-model-based approach with pseudo-relevance feedback for addressing the vocabulary problem in visual BOW mining. We employ the pseudo positive images produced in response to the original query as a set of "cues" to gradually refine the query language model. Unlike traditional approaches that only ruggedly append feedback information into the original query, the proposed approach reconstructs the query language model with finer granularities so that the query concepts can be captured more accurately. The proposed approach is experimentally evaluated using two different types of image object databases. Our algorithms are shown to bring significant improvement in the retrieval accuracy over a non-feedback baseline, and achieve better performance than conventional feedback approaches.

We aim to improve the bag-of-visual-words (BOW) model for near-duplicate image retrieval, by introducing a more fine-grained pseudo-relevance feedback process. The BOW method is based on vector quantization of affine invariant descriptors of image patches. Despite its popularity and simplicity, the retrieval performance of BOW is often unsatisfactory due to the large and diverse variations of near-duplicate images. We thus propose an information-theoretic feedback framework that employs available cues in the search result to find more relevant duplicate images which are hard to retrieve by using conventional BOW approaches. Our algorithm is experimentally evaluated under a severely attacked image database, and shown to significantly improve the retrieval accuracy over a non-feedback baseline.

Conventional image copy detection research concentrates on finding features that are robust enough to resist various kinds of image attacks. However, finding a globally effective fealure is difficult and, in many cases, domain dependent. Instead of imply extracting features from copyrighted images directly, we propose a new framework called the extended feature set for detecting copies of images. In our approach, virtual prior attacks are applied to copyrighted images to generate novel features, which serve as training data. The copy-detection problem can be solved by learning classifiers from the training data, thus, generated. Our approach can be integrated into existing copy detectors to further improve their performance. Experiment results demonstrate that the proposed approach can substantially enhance the accuracy of copy detection.

In the past, many researches on image copy detection focused on finding a feature that is robust enough for various kinds of image attacks. But it is difficult to find a globally effective feature that is appropriate for many situations. In this paper, we introduce a classification framework to this problem, instead of solving the feature-selection problem. In our approach, novel features are generated by applying virtual prior attacks to copyrighted images, and the copy-detection problem is converted to a classification one that is more robust to solve. Our approach can combine existing image copy detectors and further raise their performances.