Turn-by-turn (TBT) navigation systems are integral to modern driving experiences, providing real-time audio instructions to guide drivers safely to destinations. However, existing audio instruction policy often relies on rule-based approaches that struggle to balance informational content with cognitive load, potentially leading to driver confusion or missed turns in complex environments. To overcome these difficulties, we first model the generation of navigation instructions as a multi-task learning problem by decomposing the audio content into combinations of modular elements. Then, we propose a novel deep learning framework that leverages the powerful spatiotemporal information processing capabilities of Transformers and the strong multi-task learning abilities of Mixture of Experts (MoE) to generate real-time, context-aware audio instructions for TBT driving navigation. A cloud-edge collaborative architecture is implemented to handle the computational demands of the model, ensuring scalability and real-time performance for practical applications. Experimental results in the real world demonstrate that the proposed method significantly reduces the yaw rate (the proportion of vehicles deviating from navigation routes) compared to traditional methods, delivering clearer and more effective audio instructions. This is the first large-scale application of deep learning in driving audio navigation, marking a substantial advancement in intelligent transportation and driving assistance technologies.

In computational social systems, recommendation functionality plays a pivotal role in influencing user behavior, enhancing user experience, and driving engagement. To help recommendation functionality to better suggest relevant content or items, the social platforms usually utilize large-scale knowledge discovery techniques to analyze trends in user interactions and extract patterns from large datasets. Click-through rate (CTR) prediction is crucial in recommendation systems for measuring effectiveness, understanding user behavior, training and optimizing models, impacting business outcomes, enhancing personalization, and identifying issues. It provides actionable insights that assist in continuously refining and improving the recommendation process. Traditional deep learning-based CTR prediction models cannot work well for recommendation in social systems due to the data sparsity and the long-tail data problems since the representation learned from the user behavior is basically dominated by the major part of the data. In this article, we propose a multitask-based self-supervised learning model (MTSSL) that can better deal with sparse and long-tail user interaction data. Specifically, we first transform the CTR prediction task into the multitask joint learning framework with a set of shared subnetworks. Each subnetwork learns a representation of the entire user data, and hence, the sparse and long-tail data would have opportunity to fall into the best matched representation space of historical user behavior. Moreover, two kinds of self-supervision signals are employed to guide the learning of the representations. Extensive experiments over four user interaction datasets demonstrate the superiority of our proposed MTSSL over state-of-art models for recommendations. In terms of online A/B test, our model achieves around 3% better performance than the counterparts.

Recently around half of the world’s current population resides in urban areas and benefit from rich services in the smart city. The majority of smart city services are recommendation-related services, and with the development of Internet, most recommendation services in smart economy are online recommendations. Online travel platforms (OTPs) (like Booking, Airbnb, Ctrip, and Fliggy) provide people sufficient resources and convenient approaches to plan and enjoy their trips in smart city. Hotel recommendation is essential for the success of OTPs. However, it is more challenging compared to item recommendation in typical E-commerce scenarios (e.g., Taobao, Jd, and YouTube). The in-nature characteristics of low-frequency and high unit-price lead to more severe sparse and long-tail data distributions. Moreover, for enhancing user experience and business returns, the recommender system seeks to improve both click-through rate (CTR) and conversion rate (CVR) where the seesaw phenomenon may occur. In order to address the aforementioned shortages in hotel recommendation, a multitask learning (MTL) method with a novel flexible multilevel extraction network [denoted as flexible MTL (FMTL)] is proposed. Particularly, FMTL takes MTL into consideration in a unified representation learning framework and is divided into feature encoding and task prediction. In the feature encoding phase, we introduce a novel multirepresentation extractor with temperature-adjusted gating mechanism (T-MRE) for each task, producing more flexible representations for sparse and long-tail data. Moreover, we fuse different representations for each task with three strategies during the prediction phase and empirically demonstrate that the simple concatenation strategy is superior than other relatively complex gating approaches. Offline and live experiments with regard to both overall metrics and user group analysis based on the scarcity of user behaviors illustrate that without significantly increasing model parameters, our FMTL model outperforms substantially over several state-of-the-art models.

An appearance model is critical for most modern trackers. While numerous novel appearance models have been proposed with demonstrated success, challenges such as occlusion and drifting are still not well addressed. In this paper, we propose a novel contextual bag-of-words (CBOW) discriminative appearance model that appropriately handles drifting and occlusion. Specifically, a contextual region containing both the target and its surroundings is explored to construct a compact representation with two bags-of-words. Each word carries discriminative appearance information that is learned by Bayesian inference. An adaptive updating approach, where the background BOWs of the CBOW model acts as a "sentinel" to prevent the integration of the background appearance with the object model, is introduced to alleviate the drifting problem. Based on CBOW, visual tracking is posed within a Bayesian framework. Moreover, an explicit detection method is employed to handle severe occlusions, which further reduces drifting. Two trackers based on the same CBOW model are implemented using either handcrafted color/texture or deep convolutional features. Our trackers are evaluated based on the popular OTB50 and VOT2015 benchmarks and perform competitively against the current state of the art. In addition, they outperform two recent BOWs trackers by a large margin using the currently available figures of merit. To take into account a tracking breakdown, we propose a new figure of merit called the mean maximum-tracked-frame ratio (MTFR) that evaluates a tracker's temporal persistence without any interruption. Experiments with OTB50 demonstrate the superior robustness of our tracker compared with all other evaluated trackers on the basis of MTFR.

In this letter, we propose a novel discriminative bag-of-words (DBoW) model that can both adapt to appearance variations over time and reduce the commonly observed drifting problem in online tracking. Specifically, a contextual region containing both the object and its surroundings is explored to construct a compact representation with two bags-of-words. Each visual word is learned to carry discriminative appearance cues for the object. In order to alleviate the drifting problem, an adaptive updating approach is introduced to prevent the integration of the background into the object model. Based on DBoW model, a robust and near real-time tracker is proposed, where tracking is accomplished by searching the candidate that best matches to the maintained DBoW model. Extensive experimental results demonstrate competitive performance of the proposed method to state-of-the-art algorithms.

In this paper, we propose a fast single image dehazing algorithm using domain transformation-based edge-preserving filter and weighed quadtree subdivision. The proposed algorithm first estimates the atmospheric light in the minimal component of the haze image based on the weighed quadtree subdivision. Since the dark channel prior is invalid when the scene objects are with similar intensity to the atmospheric light, we then modify the transmission map to improve the adaptability of the algorithm. Due to the observation that haze is widely spread in the hazy image, the transmission map should be smoothly changed over the scene. The proposed algorithm utilizes the domain transformation-based edge-preserving filter to smooth the details and preserve edges and corners, aiming to obtain the refined transmission map. Experimental results demonstrate that the proposed algorithm produces comparative or even better results compared to the state-of-the-art methods but with much higher computational efficiency.

For the color fadedness and the contrast reduction of hazy images, we propose a novel method based on digital total variation (TV) filter with color transfer (DTVFCT) for single color image dehazing. The estimation of atmospheric veil is a filtering problem on the minimal component image and the digital TV filter is applied to preserve the edges and gradients of images to avoid halo artifacts. To obtain a quality vision of a resulting image, color transfer is utilized to protect the final haze-free image's color from high dynamic. The proposed method enjoys a celerity because of fast convergence of digital TV filter. The experiment results show the vivid color and the accuracy of the restored edges and gradients in the comparative study.

Visual tracking under illumination changes is a challenging task for various computer vision systems. In this paper, we propose a robust and efficient tracking algorithm based on a novel feature descriptor called joint gradient-intensity histogram, which exploits both the pixel intensity and gradient orientation features. Unlike the traditional histogram based algorithms, the joint gradient-intensity histogram is extracted efficiently by employing the integral histogram, which is calculated recursively in the specially designed local image region, making the exhaustively searching process in real time. Experiments with the state-of-art tracking methods demonstrate superior performance of the proposed tracking algorithm under illumination changes.

In this letter, we propose a robust and real-time visual tracking algorithm via a novel kernel based multiple cue adaptive appearance model (KBMCAAM). In particular, the appearance model is constructed with a naive Bayes classifier which is trained utilizing sparse multi-scale Haar-like features weighted by a spatial kernel function. Moreover, multiple image cues are integrated to improve the model's discriminative capacity. Experimental results demonstrate the superior performance of our proposed method to many state-of-art algorithms.