Yuhongze Zhou

Liguang Zhou, Yuhongze Zhou, Xiaonan Qi et al.

Audio-Visual scene understanding is a challenging problem due to the unstructured spatial-temporal relations that exist in the audio signals and spatial layouts of different objects and various texture patterns in the visual images. Recently, many studies have focused on abstracting features from convolutional neural networks while the learning of explicit semantically relevant frames of sound signals and visual images has been overlooked. To this end, we present an end-to-end framework, namely attentional graph convolutional network (AGCN), for structure-aware audio-visual scene representation. First, the spectrogram of sound and input image is processed by a backbone network for feature extraction. Then, to build multi-scale hierarchical information of input features, we utilize an attention fusion mechanism to aggregate features from multiple layers of the backbone network. Notably, to well represent the salient regions and contextual information of audio-visual inputs, the salient acoustic graph (SAG) and contextual acoustic graph (CAG), salient visual graph (SVG), and contextual visual graph (CVG) are constructed for the audio-visual scene representation. Finally, the constructed graphs pass through a graph convolutional network for structure-aware audio-visual scene recognition. Extensive experimental results on the audio, visual and audio-visual scene recognition datasets show that promising results have been achieved by the AGCN methods. Visualizing graphs on the spectrograms and images have been presented to show the effectiveness of proposed CAG/SAG and CVG/SVG that could focus on the salient and semantic relevant regions.

Liguang Zhou, Junjie Hu, Yuhongze Zhou et al.

Unbiased scene graph generation (USGG) is a challenging task that requires predicting diverse and heavily imbalanced predicates between objects in an image. To address this, we propose a novel framework peer learning that uses predicate sampling and consensus voting (PSCV) to encourage multiple peers to learn from each other. Predicate sampling divides the predicate classes into sub-distributions based on frequency, and assigns different peers to handle each sub-distribution or combinations of them. Consensus voting ensembles the peers' complementary predicate knowledge by emphasizing majority opinion and diminishing minority opinion. Experiments on Visual Genome show that PSCV outperforms previous methods and achieves a new state-of-the-art on SGCls task with 31.6 mean.

Liguang Zhou, Yuhongze Zhou, Tin Lun Lam et al.

Scene graph generation (SGG) has gained tremendous progress in recent years. However, its underlying long-tailed distribution of predicate classes is a challenging problem. For extremely unbalanced predicate distributions, existing approaches usually construct complicated context encoders to extract the intrinsic relevance of scene context to predicates and complex networks to improve the learning ability of network models for highly imbalanced predicate distributions. To address the unbiased SGG problem, we introduce a simple yet effective method dubbed Context-Aware Mixture-of-Experts (CAME) to improve model diversity and mitigate biased SGG without complicated design. Specifically, we propose to integrate the mixture of experts with a divide and ensemble strategy to remedy the severely long-tailed distribution of predicate classes, which is applicable to the majority of unbiased scene graph generators. The biased SGG is thereby reduced, and the model tends to anticipate more evenly distributed predicate predictions. To differentiate between various predicate distribution levels, experts with the same weights are not sufficiently diverse. In order to enable the network dynamically exploit the rich scene context and further boost the diversity of model, we simply use the built-in module to create a context encoder. The importance of each expert to scene context and each predicate to each expert is dynamically associated with expert weighting (EW) and predicate weighting (PW) strategy. We have conducted extensive experiments on three tasks using the Visual Genome dataset, showing that CAME outperforms recent methods and achieves state-of-the-art performance. Our code will be available publicly.

Yuhongze Zhou, Qinjie Xiao

We propose a generative framework based on generative adversarial network (GAN) to enhance facial attractiveness while preserving facial identity and high-fidelity. Given a portrait image as input, having applied gradient descent to recover a latent vector that this generative framework can use to synthesize an image resemble to the input image, beauty semantic editing manipulation on the corresponding recovered latent vector based on InterFaceGAN enables this framework to achieve facial image beautification. This paper compared our system with Beholder-GAN and our proposed result-enhanced version of Beholder-GAN. It turns out that our framework obtained state-of-art attractiveness enhancement results. The code is available at https://github.com/zoezhou1999/BeautifyBasedOnGAN.

Zhihao Shi, Dong Huo, Yuhongze Zhou et al.

Current 3D inpainting and object removal methods are largely limited to front-facing scenes, facing substantial challenges when applied to diverse, "unconstrained" scenes where the camera orientation and trajectory are unrestricted. To bridge this gap, we introduce a novel approach that produces inpainted 3D scenes with consistent visual quality and coherent underlying geometry across both front-facing and unconstrained scenes. Specifically, we propose a robust 3D inpainting pipeline that incorporates geometric priors and a multi-view refinement network trained via test-time adaptation, building on a pre-trained image inpainting model. Additionally, we develop a novel inpainting mask detection technique to derive targeted inpainting masks from object masks, boosting the performance in handling unconstrained scenes. To validate the efficacy of our approach, we create a challenging and diverse benchmark that spans a wide range of scenes. Comprehensive experiments demonstrate that our proposed method substantially outperforms existing state-of-the-art approaches.

Yuhongze Zhou, Liguang Zhou, Tin Lun Lam et al.

In recent years, self-supervised denoising methods have shown impressive performance, which circumvent painstaking collection procedure of noisy-clean image pairs in supervised denoising methods and boost denoising applicability in real world. One of well-known self-supervised denoising strategies is the blind-spot training scheme. However, a few works attempt to improve blind-spot based self-denoiser in the aspect of network architecture. In this paper, we take an intuitive view of blind-spot strategy and consider its process of using neighbor pixels to predict manipulated pixels as an inpainting process. Therefore, we propose a novel Mask Guided Residual Convolution (MGRConv) into common convolutional neural networks, e.g. U-Net, to promote blind-spot based denoising. Our MGRConv can be regarded as soft partial convolution and find a trade-off among partial convolution, learnable attention maps, and gated convolution. It enables dynamic mask learning with appropriate mask constrain. Different from partial convolution and gated convolution, it provides moderate freedom for network learning. It also avoids leveraging external learnable parameters for mask activation, unlike learnable attention maps. The experiments show that our proposed plug-and-play MGRConv can assist blind-spot based denoising network to reach promising results on both existing single-image based and dataset-based methods.

Yuhongze Zhou, Liguang Zhou, Tin Lun Lam et al.

Natural image matting aims to precisely separate foreground objects from background using alpha matte. Fully automatic natural image matting without external annotation is challenging. Well-performed matting methods usually require accurate labor-intensive handcrafted trimap as extra input, while the performance of automatic trimap generation method of dilating foreground segmentation fluctuates with segmentation quality. Therefore, we argue that how to handle trade-off of additional information input is a major issue in automatic matting. This paper presents a semantic-guided automatic natural image matting pipeline with Trimap Generation Network and light-weight non-local attention, which does not need trimap and background as input. Specifically, guided by foreground segmentation, Trimap Generation Network estimates accurate trimap. Then, with estimated trimap as guidance, our light-weight Non-local Matting Network with Refinement produces final alpha matte, whose trimap-guided global aggregation attention block is equipped with stride downsampling convolution, reducing computation complexity and promoting performance. Experimental results show that our matting algorithm has competitive performance with state-of-the-art methods in both trimap-free and trimap-needed aspects.