Video diffusion models have recently shown strong capability in synthesizing high-fidelity videos in various ways, including prediction, interpolation, and unconditional generation. However, their synthesis ability credits a lot to leveraging large denoising models to reverse the long noise-adding process, which also brings extremely expansive sampling and training costs. After examining the source of the computation cost, we confirm that the main calculation comes from the redundancy of the convolution. To address this issue, we propose Efficiency-optimized Video Diffusion Models to reduce the network's computation cost by minimizing the input and output channels of the convolution. First, a bottleneck residual pathway is proposed to conduct a channel-wise downsample to the convolution pathways, which extracts crucial information from the input and reduces computation cost. Second, a three-path channel split strategy is proposed to reduce channel redundancy by handling part of the input channels with more efficient pointwise convolution and skip-connection pathways. Furthermore, a mixed self-attention mechanism is proposed to optimize the computation cost of the self-attention in the network by adaptively choosing the algorithm with lower time complexity according to the input token lengths and hidden dimensions. Extensive experiments on three downstream tasks show that our Efficiency-optimized Video Diffusion Models can achieve a 10x speed-up while achieving comparable or even better results in the performance of fidelity compared with the state-of-the-art methods. The code is available at https://github.com/PKU-ICST-MIPL/EVDM_ACMMM2023.

In this paper, we present a Motion-aware Video Diffusion Model (MV-Diffusion) for enhancing the temporal consistency of generated videos using autoregressive diffusion models. Despite the success of diffusion models in various vision generation tasks, generating high-quality and realistic videos with coherent temporal structure remains a challenging problem. Current methods have primarily focused on capturing implicit motion features within a restricted window of RGB frames, rather than explicitly modeling the motion. To address this, we focus on improving the temporal modeling ability of the current autoregressive video diffusion approach by leveraging rich temporal trajectory information in a global context and explicitly modeling local motion trends. The main contributions of this research include: (1) a Trajectory Modeling (TM) block that enhances the model's conditioning by incorporating global motion trajectory information, (2) a Motion Trend Attention (MTA) block that utilizes a cross-attention mechanism to explicitly infer motion trends from the optical flow rather than implicitly learning from RGB input. Experimental results on three video generation tasks using four datasets show the effectiveness of our proposed MV-Diffusion, outperforming existing state-of-the-art approaches. The code is available at https://github.com/PKU-ICST-MIPL/MV-Diffusion_ACMMM2023.

Camouflaged object detection is a challenging task that aims to identify objects having similar texture to the surroundings. This paper presents to amplify the subtle texture difference between camouflaged objects and the background for camouflaged object detection by formulating multiple texture-aware refinement modules to learn the texture-aware features in a deep convolutional neural network. The texture-aware refinement module computes the biased co-variance matrices of feature responses to extract the texture information, adopts an affinity loss to learn a set of parameter maps that help to separate the texture between camouflaged objects and the background, and leverages a boundary-consistency loss to explore the structures of object details. We evaluate our network on the benchmark datasets for camouflaged object detection both qualitatively and quantitatively. Experimental results show that our approach outperforms various state-of-the-art methods by a large margin.

Text-to-image generation aims to generate images from text descriptions. Its main challenge lies in two aspects: (1) Semantic consistency, i.e., the generated images should be semantically consistent with the input text; and (2) Visual reality, i.e., the generated images should look like real images. To ensure text-image consistency, existing works mainly learn to establish the cross-modal representations via a text encoder and image encoder. However, due to the limited representation capability of the fixed-length embeddings and the flexibility of the free-form text descriptions, the learned text-to-image model is incapable of maintaining the semantic consistency between image local regions and fine-grained descriptions. As a result, the generated images sometimes miss some fine-grained attributes of the generated object, such as the color or shape of a part of the object. To address this issue, this paper proposes a Local Feature Refinement Based Generative Adversarial Network (LFR-GAN) , which first divides the text into some independent fine-grained attributes and generates an initial image, then refines the image details based on these attributes. The main contributions are three-fold: (1) An attribute modeling approach is proposed to model the fine-grained text descriptions by mapping them into representations of independent attributes, which provides more fine-grained details for image generation. (2) A local feature refinement approach is proposed to enable the generated image to form a complete reflection of the fine-grained attributes contained in the text description. (3) A multi-stage generation approach is proposed to realize the fine-grained manipulation of complex images progressively, which aims to improve the performance of the refinement and generate photo-realistic images. Extensive experiments on the CUB and Oxford102 datasets show the effectiveness of our LFR-GAN approach in both text-to-image generation and text-guided image manipulation tasks. Our LFR-GAN approach shows superior performance compared to the state-of-the-art methods. The codes will be released at https://github.com/PKU-ICST-MIPL/LFR-GAN_TOMM2023 .

This work presents a gated non-local deep residual learning framework for image deraining. It can avoid the over-deraining or under-deraining caused by the global residual learning in existing deraining networks, since the learned soft gate in our method adaptively adjusts the amount of global residual to be passed for generating the final derained result. To generate feature maps for global residual prediction, we develop a non-local guided attention module (NLAM), which first obtains non-local features by exploiting spatial inter-dependencies among all the feature positions of local features produced by convolutional neural network (CNN), and then leverages the attention mechanism to merge the local and non-local features based on their complementary relation. Moreover, we develop a channel-wise gated prediction module to learn a soft gate on the global residual by explicitly modelling channel inter-dependencies of the feature maps obtained from NLAM. Experiments on four deraining benchmark datasets and real-world rainy images show that our network has a quantitative and qualitative improvement over state-of-the-arts.

This paper presents a deep multi-model fusion network to attentively integrate multiple models to separate layers and boost the performance in single-image dehazing. To do so, we first formulate the attentional feature integration module to maximize the integration of the convolutional neural network (CNN) features at different CNN layers and generate the attentional multi-level integrated features (AMLIF). Then, from the AMLIF, we further predict a haze-free result for an atmospheric scattering model, as well as for four haze-layer separation models, and then fuse the results together to produce the final haze-free image. To evaluate the effectiveness of our method, we compare our network with several state-of-the-art methods on two widely-used dehazing benchmark datasets, as well as on two sets of real-world hazy images. Experimental results demonstrate clear quantitative and qualitative improvements of our method over the state-of-the-arts.

Saliency detection is a fundamental yet challenging task in computer vision, aiming at highlighting the most visually distinctive objects in an image. We propose a novel recurrent residual refinement network (R^3Net) equipped with residual refinement blocks (RRBs) to more accurately detect salient regions of an input image. Our RRBs learn the residual between the intermediate saliency prediction and the ground truth by alternatively leveraging the low-level integrated features and the high-level integrated features of a fully convolutional network (FCN). While the low-level integrated features are capable of capturing more saliency details, the high-level integrated features can reduce non-salient regions in the intermediate prediction. Furthermore, the RRBs can obtain complementary saliency information of the intermediate prediction, and add the residual into the intermediate prediction to refine the saliency maps. We evaluate the proposed R^3Net on five widely-used saliency detection benchmarks by comparing it with 16 state-of-the-art saliency detectors. Experimental results show that our network outperforms our competitors in all the benchmark datasets.