Blind face restoration is a challenging task due to the unknown and complex degradation. Although face prior-based methods and reference-based methods have recently demonstrated high-quality results, the restored images tend to contain over-smoothed results and lose identity-preserved details when the degradation is severe. It is observed that this is attributed to short-range dependencies, the intrinsic limitation of convolutional neural networks. To model long-range dependencies, we propose a Transformer-based blind face restoration method, named BFRFormer, to reconstruct images with more identity-preserved details in an end-to-end manner. In BFRFormer, to remove blocking artifacts, the wavelet discriminator and aggregated attention module are developed, and spectral normalization and balanced consistency regulation are adaptively applied to address the training instability and over-fitting problem, respectively. Extensive experiments show that our method outperforms state-of-the-art methods on a synthetic dataset and four real-world datasets. The source code, Casia-Test dataset, and pre-trained models is released at https://github.com/s8Znk/BFRFormer.

While the recently introduced Tree of Thoughts (ToT) has heralded advancements in allowing Large Language Models (LLMs) to reason through foresight and backtracking for global decision-making, it has overlooked the inherent local uncertainties in intermediate decision points or "thoughts". These local uncertainties, intrinsic to LLMs given their potential for diverse responses, remain a significant concern in the reasoning process. Addressing this pivotal gap, we introduce the Tree of Uncertain Thoughts (TouT) — a reasoning framework tailored for LLMs. Our TouT effectively leverages Monte Carlo Dropout to quantify uncertainty scores associated with LLMs’ diverse local responses at these intermediate steps. By marrying this local uncertainty quantification with global search algorithms, TouT enhances the model’s precision in response generation. We substantiate our approach with rigorous experiments on two demanding planning tasks: Game of 24 and Mini Crosswords. The empirical evidence underscores TouT’s superiority over both ToT and chain-of-thought prompting methods.

Human pose forecasting that aims to predict the body poses happening in the future is an important task in computer vision. However, long-term pose forecasting is particularly challenging because modeling long-range dependencies across the spatial-temporal level is hard for joint-based representation. Another challenge is uncertainty prediction since the future prediction is not a deterministic process. In this article, we present a novelBayesianSpatial-TemporalGraphTransformer (BSTG-Trans) for predicting accurate, diverse, and uncertain future poses. First, we apply a spatial-temporal graph transformer as an encoder and a temporal-spatial graph transformer as a decoder for modeling the long-range spatial-temporal dependencies across pose joints to generate the long-term future body poses. Furthermore, we propose a Bayesian sampling module for uncertainty quantization of diverse future poses. Finally, a novel uncertainty estimation metric, namely Uncertainty Absolute Error is introduced for measuring both the accuracy and uncertainty of each predicted future pose. We achieve state-of-the-art performance against other baselines on Human3.6 M and HumanEva-I in terms of accuracy, diversity, and uncertainty for long-term pose forecasting. Moreover, our comprehensive ablation studies demonstrate the effectiveness and generalization of each module proposed in our BSTG-Trans.

We observed an attribute entanglement phenomenon: samples with similar attributes but from different classes can easily result in recognition errors. This problem is an important cause that results in recognition errors. To address this problem, we propose a new loss function, namely the entanglement loss. It penalizes the compactness between the misclassified entangled samples and their misclassified class centers, such that the features of entangled samples are apart from the misclassified classes. The proposed loss function can effectively enhance the discriminative power of the deeply learned features, thus recognition performance can be significantly improved. Experimental results show that our method outperforms the previous state-of-the-art methods on PASCAL VOC 2012 Action and ASLAN datasets.

Summary Selective maintenance is often applied in many industrial environments where maintenance is performed between sequence missions. When the mission time is stochastic and there are multiple maintenance workers with different capacities, the system reliability of the next work mission can be maximized by using a stochastic model under the constraint of the limit maintenance time. The optimal maintenance strategy is obtained with an optimization algorithm. A simulation was performed to verify the validity and feasibility of the proposed model.

Kernel methods have achieved tremendous success in the past two decades. In the current big data era, data collection has grown tremendously. However, existing kernel methods are not scalable enough both at the training and predicting steps. To address this challenge, in this paper, we first introduce a general sparse kernel learning formulation based on the random feature approximation, where the loss functions are possibly non-convex. Then we propose a new asynchronous parallel doubly stochastic algorithm for large scale sparse kernel learning (AsyDSSKL). To the best our knowledge, AsyDSSKL is the first algorithm with the techniques of asynchronous parallel computation and doubly stochastic optimization. We also provide a comprehensive convergence guarantee to AsyDSSKL. Importantly, the experimental results on various large-scale real-world datasets show that, our AsyDSSKL method has the significant superiority on the computational efficiency at the training and predicting steps over the existing kernel methods.

Action recognition in still images is a challenging task in computer vision. Recent successes in deep feature-learning advance this research, employing robust and rich-semantic feature representation. However, the issue that recognition fails when two action images share similar contexts is long-standing. In this paper, we employ metric learning method to address within-class and between-class confusions in action recognition. We propose a novel loss function, named composite-triplet loss. Supervised by this loss function, our method directly learns a similarity function from data. Employing a customized human action and context detection network, we obtain highly discriminative action image embeddings, which can be used in action image recognition and other tasks. Our approach is evaluated on the still-image action recognition task and the image caption generation task. On the PASCAL VOC dataset, our approach outperforms state-of-the-art methods, achieving 90.6% mean AP.

Visual context is fundamental to understand human actions in videos. However, to efficiently employ temporal context information presents an enormous challenge to this area. Two main problems are long-standing: (1) video frames are redundant while discriminative information is sparse; (2) large amount of interference information is mixed in frame sequences. These factors results in redundant computation and recognition failures. In this paper, we propose a learnable temporal attention mechanism to automatically select important time points from action sequences. We design an unsupervised Recurrent Temporal Sparse Autoencoder (RTSAE) network, which learns to extract sparse key-frames to sharpen discriminative yet to retain descriptive capability, as well to shield interfere information. By applying this technique to a recent proposed action recognition model Adaptive Recurrent-convolutional Hybrid network (ARCH), we significantly improve its performance in both speed and accuracy. Experiments demonstrate that, with the help of the RTSAE, ARCH outperforms most state-of-the-art methods on UCF101 and HMDB51 datasets.