Conditioned diffusion models have demonstrated state-of-the-art text-to-image synthesis capacity. Recently, most works focus on synthesizing independent images; While for real-world applications, it is common and necessary to generate a series of coherent images for story-stelling. In this work, we mainly focus on story visualization and continuation tasks and propose AR-LDM, a latent diffusion model auto-regressively conditioned on history captions and generated images. Moreover, AR-LDM can generalize to new characters through adaptation. To our best knowledge, this is the first work successfully leveraging diffusion models for coherent visual story synthesizing. It also extends the text-conditioned method to multimodal conditioning. Quantitative results show that AR-LDM achieves SoTA FID scores on PororoSV, FlintstonesSV, and the adopted challenging dataset VIST containing natural images. Large-scale human evaluations show that AR-LDM has superior performance in terms of quality, relevance, and consistency. Code available at this https URL

Partial person re-identification (ReID) aims to solve the problem of image spatial misalignment due to occlusions or out-of-views. Despite significant progress through the introduction of additional information, such as human pose landmarks, mask maps, and spatial information, partial person ReID remains challenging due to noisy keypoints and impressionable pedestrian representations. To address these issues, we propose a unified attribute-guided collaborative learning scheme for partial person ReID. Specifically, we introduce an adaptive threshold-guided masked graph convolutional network that can dynamically remove untrustworthy edges to suppress the diffusion of noisy keypoints. Furthermore, we incorporate human attributes and devise a cyclic heterogeneous graph convolutional network to effectively fuse cross-modal pedestrian information through intra- and inter-graph interaction, resulting in robust pedestrian representations. Finally, to enhance keypoint representation learning, we design a novel part-based similarity constraint based on the axisymmetric characteristic of the human body. Extensive experiments on multiple public datasets have shown that our model achieves superior performance compared to other state-of-the-art baselines.

Humans are capable of accumulating knowledge by sequentially learning different tasks, while neural networks fail to achieve this due to catastrophic forgetting problems. Most current incremental learning methods focus more on tackling catastrophic forgetting for traditional classification networks. Notably, however, embedding networks that are basic architectures for many metric learning applications also suffer from this problem. Moreover, the most significant difficulty for continual embedding networks is that the relationships between the latent features and prototypes of previous tasks will be destroyed once new tasks have been learned. Accordingly, we propose a novel incremental method for embedding networks, called the disentangled representation translation (DRT) method, to obtain the discriminative class-disentangled features without reusing any samples of previous tasks and while avoiding the perturbation of task-related information. Next, a mask-guided module is specifically explored to adaptively change or retain the valuable information of latent features. This module enables us to effectively preserve the discriminative yet representative features in the disentangled translation process. In addition, DRT can easily be equipped with a regularization item of incremental learning to further improve performance. We conduct extensive experiments on four popular datasets; as the experimental results clearly demonstrate, our method can effectively alleviate the catastrophic forgetting problem for embedding networks.

Words are treated as atomic units in natural language processing tasks and it is a fundamental step to represent them as vectors for supporting subsequent computations. GloVe is a widely used machine learning model to train word vectors. Generally, a large corpus and high computation resources are required to train high-quality word vectors using GloVe, making it difficult for users to train their own word vectors by themselves. A natural choice nowadays is to outsource the training process to the cloud. However, coming with such cloud-based training services are serious privacy concerns, which should be well addressed. In this paper, we design, implement, and evaluate PPGloVe, the first system framework that supports privacy-preserving word vectors training using GloVe over encrypted data of multiple participants. We first decompose the training task and show that previous privacy-preserving machine learning techniques are not practical for this task. We then construct a new secure training strategy to delicately bridge lightweight cryptographic techniques with GloVe in depth to support privacy-preserving GloVe training on the cloud. By design, the corpora of the participants and the trained word vectors are kept private along the whole training process. Extensive experiments over three datasets of different scales demonstrate that PPGloVe produces word vectors with promising quality comparable to plaintext training, with practically affordable overhead.