Universal style transfer (UST) infuses styles from arbitrary reference images into content images. Existing methods, while enjoying many practical successes, are unable of explaining experimental observations, including different performances of UST algorithms in preserving the spatial structure of content images. In addition, methods are limited to cumbersome global controls on stylization, so that they require additional spatial masks for desired stylization. In this work, we provide a systematic Fourier analysis on a general framework for UST. We present an equivalent form of the framework in the frequency domain. The form implies that existing algorithms treat all frequency components and pixels of feature maps equally, except for the zero-frequency component. We connect Fourier amplitude and phase with Gram matrices and a content reconstruction loss in style transfer, respectively. Based on such equivalence and connections, we can thus interpret different structure preservation behaviors between algorithms with Fourier phase. Given the interpretations we have, we propose two manipulations in practice for structure preservation and desired stylization. Both qualitative and quantitative experiments demonstrate the competitive performance of our method against the state-of-the-art methods. We also conduct experiments to demonstrate (1) the abovementioned equivalence, (2) the interpretability based on Fourier amplitude and phase and (3) the controllability associated with frequency components.
In surface defect detection, due to the extreme imbalance in the number of positive and negative samples, positive-samples-based anomaly detection methods have received more and more attention. Specifically, reconstruction-based methods are the most popular. However, exiting methods are either difficult to repair abnormal foregrounds or reconstruct clear backgrounds. Therefore, we propose a clear memory-augmented auto-encoder. At first, we propose a novel clear memory-augmented module, which combines the encoding and memory-encoding in a way of forgetting and inputting, thereby repairing abnormal foregrounds and preservation clear backgrounds. Secondly, a general artificial anomaly generation algorithm is proposed to simulate anomalies that are as realistic and feature-rich as possible. At last, we propose a novel multi scale feature residual detection method for defect segmentation, which makes the defect location more accurate. CMA-AE conducts comparative experiments using 11 state-of-the-art methods on five benchmark datasets, showing an average 18.6% average improvement in F1-measure.
Previous works on human motion prediction follow the pattern of building a mapping relation between the sequence observed and the one to be predicted. However, due to the inherent complexity of multivariate time series data, it still remains a challenge to find the extrapolation relation between motion sequences. In this paper, we present a new prediction pattern, which introduces previously overlooked human poses, to implement the prediction task from the view of interpolation. These poses exist after the predicted sequence, and form the privileged sequence. To be specific, we first propose an InTerPolation learning Network (ITP-Network) that encodes both the observed sequence and the privileged sequence to interpolate the in-between predicted sequence, wherein the embedded Privileged-sequence-Encoder (Priv-Encoder) learns the privileged knowledge (PK) simultaneously. Then, we propose a Final Prediction Network (FP-Network) for which the privileged sequence is not observable, but is equipped with a novel PK-Simulator that distills PK learned from the previous network. This simulator takes as input the observed sequence, but approximates the behavior of Priv-Encoder, enabling FP-Network to imitate the interpolation process. Extensive experimental results demonstrate that our prediction pattern achieves state-of-the-art performance on benchmarked H3.6M, CMU-Mocap and 3DPW datasets in both short-term and long-term predictions.
Data-Efficient GANs (DE-GANs), which aim to learn generative models with a limited amount of training data, encounter several challenges for generating high-quality samples. Since data augmentation strategies have largely alleviated the training instability, how to further improve the generative performance of DE-GANs becomes a hotspot. Recently, contrastive learning has shown the great potential of increasing the synthesis quality of DE-GANs, yet related principles are not well explored. In this paper, we revisit and compare different contrastive learning strategies in DE-GANs, and identify (i) the current bottleneck of generative performance is the discontinuity of latent space; (ii) compared to other contrastive learning strategies, Instance-perturbation works towards latent space continuity, which brings the major improvement to DE-GANs. Based on these observations, we propose FakeCLR, which only applies contrastive learning on perturbed fake samples, and devises three related training techniques: Noise-related Latent Augmentation, Diversity-aware Queue, and Forgetting Factor of Queue. Our experimental results manifest the new state of the arts on both few-shot generation and limited-data generation. On multiple datasets, FakeCLR acquires more than 15% FID improvement compared to existing DE-GANs. Code is available at https://github.com/iceli1007/FakeCLR.
For neural video codec, it is critical, yet challenging, to design an efficient entropy model which can accurately predict the probability distribution of the quantized latent representation. However, most existing video codecs directly use the ready-made entropy model from image codec to encode the residual or motion, and do not fully leverage the spatial-temporal characteristics in video. To this end, this paper proposes a powerful entropy model which efficiently captures both spatial and temporal dependencies. In particular, we introduce the latent prior which exploits the correlation among the latent representation to squeeze the temporal redundancy. Meanwhile, the dual spatial prior is proposed to reduce the spatial redundancy in a parallel-friendly manner. In addition, our entropy model is also versatile. Besides estimating the probability distribution, our entropy model also generates the quantization step at spatial-channel-wise. This content-adaptive quantization mechanism not only helps our codec achieve the smooth rate adjustment in single model but also improves the final rate-distortion performance by dynamic bit allocation. Experimental results show that, powered by the proposed entropy model, our neural codec can achieve 18.2% bitrate saving on UVG dataset when compared with H.266 (VTM) using the highest compression ratio configuration. It makes a new milestone in the development of neural video codec. The codes are at https://github.com/microsoft/DCVC.
Unmanned aerial vehicles (UAVs) are foreseen to constitute promising airborne communication devices as a benefit of their superior channel quality. But UAV-to-ground (U2G) communications are vulnerable to eavesdropping. Hence, we conceive a sophisticated physical layer security solution for improving the secrecy rate of multi-antenna aided U2G systems. Explicitly, the secrecy rate of the U2G MIMO wiretap channels is derived by using random matrix theory. The resultant explicit expression is then applied in the joint optimization of the MIMO transceiver and the UAV location relying on an alternating optimization technique. Our numerical results show that the joint transceiver and location optimization conceived facilitates secure communications even in the challenging scenario, where the legitimate channel of confidential information is inferior to the eavesdropping channel.
This paper introduces the schemes of Team LingJing's experiments in NLPCC-2022-Shared-Task-4 Multi-modal Dialogue Understanding and Generation (MDUG). The MDUG task can be divided into two phases: multi-modal context understanding and response generation. To fully leverage the visual information for both scene understanding and dialogue generation, we propose the scene-aware prompt for the MDUG task. Specifically, we utilize the multi-tasking strategy for jointly modelling the scene- and session- multi-modal understanding. The visual captions are adopted to aware the scene information, while the fixed-type templated prompt based on the scene- and session-aware labels are used to further improve the dialogue generation performance. Extensive experimental results show that the proposed method has achieved state-of-the-art (SOTA) performance compared with other competitive methods, where we rank the 1-st in all three subtasks in this MDUG competition.
Detection models trained by one party (server) may face severe performance degradation when distributed to other users (clients). For example, in autonomous driving scenarios, different driving environments may bring obvious domain shifts, which lead to biases in model predictions. Federated learning that has emerged in recent years can enable multi-party collaborative training without leaking client data. In this paper, we focus on a special cross-domain scenario where the server contains large-scale data and multiple clients only contain a small amount of data; meanwhile, there exist differences in data distributions among the clients. In this case, traditional federated learning techniques cannot take into account the learning of both the global knowledge of all participants and the personalized knowledge of a specific client. To make up for this limitation, we propose a cross-domain federated object detection framework, named FedOD. In order to learn both the global knowledge and the personalized knowledge in different domains, the proposed framework first performs the federated training to obtain a public global aggregated model through multi-teacher distillation, and sends the aggregated model back to each client for finetuning its personalized local model. After very few rounds of communication, on each client we can perform weighted ensemble inference on the public global model and the personalized local model. With the ensemble, the generalization performance of the client-side model can outperform a single model with the same parameter scale. We establish a federated object detection dataset which has significant background differences and instance differences based on multiple public autonomous driving datasets, and then conduct extensive experiments on the dataset. The experimental results validate the effectiveness of the proposed method.
The development of reconfigurable intelligent surface (RIS) has recently advanced the research of physical layer security (PLS). Beneficial impact of RIS includes but is not limited to offering a new domain of freedom (DoF) for key-less PLS optimization, and increasing channel randomness for physical layer secret key generation (PL-SKG). However, there is a lack of research studying how adversarial RIS can be used to damage the communication confidentiality. In this work, we show how a Eve controlled adversarial RIS (Eve-RIS) can be used to reconstruct the shared PLS secret key between legitimate users (Alice and Bob). This is achieved by Eve-RIS overlaying the legitimate channel with an artificial random and reciprocal channel. The resulting Eve-RIS corrupted channel enable Eve to successfully attack the PL-SKG process. To operationalize this novel concept, we design Eve-RIS schemes against two PL-SKG techniques used: (i) the channel estimation based PL-SKG, and (ii) the two-way cross multiplication based PL-SKG. Our results show a high key match rate between the designed Eve-RIS and the legitimate users. We also present theoretical key match rate between Eve-RIS and legitimate users. Our novel scheme is different from the existing spoofing-Eve, in that the latter can be easily detected by comparing the channel estimation results of the legitimate users. Indeed, our proposed Eve-RIS can maintain the legitimate channel reciprocity, which makes detection challenging. This means the novel Eve-RIS provides a new eavesdropping threat on PL-SKG, which can spur new research areas to counter adversarial RIS attacks.