Yue Han

Yue Han, Jiangning Zhang, Yabiao Wang et al.

Few-Shot Instance Segmentation (FSIS) requires detecting and segmenting novel classes with limited support examples. Existing methods based on Region Proposal Networks (RPNs) face two issues: 1) Overfitting suppresses novel class objects; 2) Dual-branch models require complex spatial correlation strategies to prevent spatial information loss when generating class prototypes. We introduce a unified framework, Reference Twice (RefT), to exploit the relationship between support and query features for FSIS and related tasks. Our three main contributions are: 1) A novel transformer-based baseline that avoids overfitting, offering a new direction for FSIS; 2) Demonstrating that support object queries encode key factors after base training, allowing query features to be enhanced twice at both feature and query levels using simple cross-attention, thus avoiding complex spatial correlation interaction; 3) Introducing a class-enhanced base knowledge distillation loss to address the issue of DETR-like models struggling with incremental settings due to the input projection layer, enabling easy extension to incremental FSIS. Extensive experimental evaluations on the COCO dataset under three FSIS settings demonstrate that our method performs favorably against existing approaches across different shots, \eg, $+8.2/+9.4$ performance gain over state-of-the-art methods with 10/30-shots. Source code and models will be available at https://github.com/hanyue1648/RefT.

Yue Han, Sri Kalyan Yarlagadda, Tonmoy Ghosh et al.

Diet is an important aspect of our health. Good dietary habits can contribute to the prevention of many diseases and improve the overall quality of life. To better understand the relationship between diet and health, image-based dietary assessment systems have been developed to collect dietary information. We introduce the Automatic Ingestion Monitor (AIM), a device that can be attached to one's eye glasses. It provides an automated hands-free approach to capture eating scene images. While AIM has several advantages, images captured by the AIM are sometimes blurry. Blurry images can significantly degrade the performance of food image analysis such as food detection. In this paper, we propose an approach to pre-process images collected by the AIM imaging sensor by rejecting extremely blurry images to improve the performance of food detection.

Weikang Wang, Guanhua Chen, Hanqing Wang et al.

Multilingual pretrained language models (mPLMs) have shown their effectiveness in multilingual word alignment induction. However, these methods usually start from mBERT or XLM-R. In this paper, we investigate whether multilingual sentence Transformer LaBSE is a strong multilingual word aligner. This idea is non-trivial as LaBSE is trained to learn language-agnostic sentence-level embeddings, while the alignment extraction task requires the more fine-grained word-level embeddings to be language-agnostic. We demonstrate that the vanilla LaBSE outperforms other mPLMs currently used in the alignment task, and then propose to finetune LaBSE on parallel corpus for further improvement. Experiment results on seven language pairs show that our best aligner outperforms previous state-of-the-art models of all varieties. In addition, our aligner supports different language pairs in a single model, and even achieves new state-of-the-art on zero-shot language pairs that does not appear in the finetuning process.

Wenjin Fu, Yue Han, Jiangpeng He et al.

Generative Adversarial Networks (GAN) have been widely investigated for image synthesis based on their powerful representation learning ability. In this work, we explore the StyleGAN and its application of synthetic food image generation. Despite the impressive performance of GAN for natural image generation, food images suffer from high intra-class diversity and inter-class similarity, resulting in overfitting and visual artifacts for synthetic images. Therefore, we aim to explore the capability and improve the performance of GAN methods for food image generation. Specifically, we first choose StyleGAN3 as the baseline method to generate synthetic food images and analyze the performance. Then, we identify two issues that can cause performance degradation on food images during the training phase: (1) inter-class feature entanglement during multi-food classes training and (2) loss of high-resolution detail during image downsampling. To address both issues, we propose to train one food category at a time to avoid feature entanglement and leverage image patches cropped from high-resolution datasets to retain fine details. We evaluate our method on the Food-101 dataset and show improved quality of generated synthetic food images compared with the baseline. Finally, we demonstrate the great potential of improving the performance of downstream tasks, such as food image classification by including high-quality synthetic training samples in the data augmentation.

Yue Han, Christopher Jerrett, Elliot Anshelevich

We study the classic facility location setting, where we are given $n$ clients and $m$ possible facility locations in some arbitrary metric space, and want to choose a location to build a facility. The exact same setting also arises in spatial social choice, where voters are the clients and the goal is to choose a candidate or outcome, with the distance from a voter to an outcome representing the cost of this outcome for the voter (e.g., based on their ideological differences). Unlike most previous work, we do not focus on a single objective to optimize (e.g., the total distance from clients to the facility, or the maximum distance, etc.), but instead attempt to optimize several different objectives simultaneously. More specifically, we consider the $l$-centrum family of objectives, which includes the total distance, max distance, and many others. We present tight bounds on how well any pair of such objectives (e.g., max and sum) can be simultaneously approximated compared to their optimum outcomes. In particular, we show that for any such pair of objectives, it is always possible to choose an outcome which simultaneously approximates both objectives within a factor of $1+\sqrt{2}$, and give a precise characterization of how this factor improves as the two objectives being optimized become more similar. For $q>2$ different centrum objectives, we show that it is always possible to approximate all $q$ of these objectives within a small constant, and that this constant approaches 3 as $q\rightarrow \infty$. Our results show that when optimizing only a few simultaneous objectives, it is always possible to form an outcome which is a significantly better than 3 approximation for all of these objectives.

Yue Han, Junwei Zhu, Yuxiang Feng et al.

Current diffusion-based face animation methods generally adopt a ReferenceNet (a copy of U-Net) and a large amount of curated self-acquired data to learn appearance features, as robust appearance features are vital for ensuring temporal stability. However, when trained on public datasets, the results often exhibit a noticeable performance gap in image quality and temporal consistency. To address this issue, we meticulously examine the essential appearance features in the facial animation tasks, which include motion-agnostic (e.g., clothing, background) and motion-related (e.g., facial details) texture components, along with high-level discriminative identity features. Drawing from this analysis, we introduce a Motion-Identity Modulated Appearance Learning Module (MIA) that modulates CLIP features at both motion and identity levels. Additionally, to tackle the semantic/ color discontinuities between clips, we design an Inter-clip Affinity Learning Module (ICA) to model temporal relationships across clips. Our method achieves precise facial motion control (i.e., expressions and gaze), faithful identity preservation, and generates animation videos that maintain both intra/inter-clip temporal consistency. Moreover, it easily adapts to various modalities of driving sources. Extensive experiments demonstrate the superiority of our method.

Yue Han, Jiangning Zhang, Junwei Zhu et al.

This work presents FaceX framework, a novel facial generalist model capable of handling diverse facial tasks simultaneously. To achieve this goal, we initially formulate a unified facial representation for a broad spectrum of facial editing tasks, which macroscopically decomposes a face into fundamental identity, intra-personal variation, and environmental factors. Based on this, we introduce Facial Omni-Representation Decomposing (FORD) for seamless manipulation of various facial components, microscopically decomposing the core aspects of most facial editing tasks. Furthermore, by leveraging the prior of a pretrained StableDiffusion (SD) to enhance generation quality and accelerate training, we design Facial Omni-Representation Steering (FORS) to first assemble unified facial representations and then effectively steer the SD-aware generation process by the efficient Facial Representation Controller (FRC). %Without any additional features, Our versatile FaceX achieves competitive performance compared to elaborate task-specific models on popular facial editing tasks. Full codes and models will be available at https://github.com/diffusion-facex/FaceX.

Meng-Xun Li, Wen-Hui Deng, Zhi-Xing Wu et al.

Vision-Language Models (VLMs) have demonstrated significant potential in medical image analysis, yet their application in intraoral photography remains largely underexplored due to the lack of fine-grained, annotated datasets and comprehensive benchmarks. To address this, we present MetaDent, a comprehensive resource that includes (1) a novel and large-scale dentistry image dataset collected from clinical, public, and web sources; (2) a semi-structured annotation framework designed to capture the hierarchical and clinically nuanced nature of dental photography; and (3) comprehensive benchmark suites for evaluating state-of-the-art VLMs on clinical image understanding. Our labeling approach combines a high-level image summary with point-by-point, free-text descriptions of abnormalities. This method enables rich, scalable, and task-agnostic representations. We curated 60,669 dental images from diverse sources and annotated a representative subset of 2,588 images using this meta-labeling scheme. Leveraging Large Language Models (LLMs), we derive standardized benchmarks: approximately 15K Visual Question Answering (VQA) pairs and an 18-class multi-label classification dataset, which we validated with human review and error analysis to justify that the LLM-driven transition reliably preserves fidelity and semantic accuracy. We then evaluate state-of-the-art VLMs across VQA, classification, and image captioning tasks. Quantitative results reveal that even the most advanced models struggle with a fine-grained understanding of intraoral scenes, achieving moderate accuracy and producing inconsistent or incomplete descriptions in image captioning. We publicly release our dataset, annotations, and tools to foster reproducible research and accelerate the development of vision-language systems for dental applications.

Shaswat Patel, Vishvesh Trivedi, Yue Han et al.

Recent work has identified a subset of attention heads in Transformer as retrieval heads, which are responsible for retrieving information from the context. In this work, we first investigate retrieval heads in multilingual contexts. In multilingual language models, we find that retrieval heads are often shared across multiple languages. Expanding the study to cross-lingual setting, we identify Retrieval-Transition heads(RTH), which govern the transition to specific target-language output. Our experiments reveal that RTHs are distinct from retrieval heads and more vital for Chain-of-Thought reasoning in multilingual LLMs. Across four multilingual benchmarks (MMLU-ProX, MGSM, MLQA, and XQuaD) and two model families (Qwen-2.5 and Llama-3.1), we demonstrate that masking RTH induces bigger performance drop than masking Retrieval Heads (RH). Our work advances understanding of multilingual LMs by isolating the attention heads responsible for mapping to target languages.

Xintian Shen, Jiangning Zhang, Jun Chen et al.

Image harmonization aims to solve the visual inconsistency problem in composited images by adaptively adjusting the foreground pixels with the background as references. Existing methods employ local color transformation or region matching between foreground and background, which neglects powerful proximity prior and independently distinguishes fore-/back-ground as a whole part for harmonization. As a result, they still show a limited performance across varied foreground objects and scenes. To address this issue, we propose a novel Global-aware Kernel Network (GKNet) to harmonize local regions with comprehensive consideration of long-distance background references. Specifically, GKNet includes two parts, \ie, harmony kernel prediction and harmony kernel modulation branches. The former includes a Long-distance Reference Extractor (LRE) to obtain long-distance context and Kernel Prediction Blocks (KPB) to predict multi-level harmony kernels by fusing global information with local features. To achieve this goal, a novel Selective Correlation Fusion (SCF) module is proposed to better select relevant long-distance background references for local harmonization. The latter employs the predicted kernels to harmonize foreground regions with both local and global awareness. Abundant experiments demonstrate the superiority of our method for image harmonization over state-of-the-art methods, \eg, achieving 39.53dB PSNR that surpasses the best counterpart by +0.78dB $\uparrow$; decreasing fMSE/MSE by 11.5\%$\downarrow$/6.7\%$\downarrow$ compared with the SoTA method. Code will be available at \href{https://github.com/XintianShen/GKNet}{here}.

Yue Han, Junwei Zhu, Keke He et al.

Current face reenactment and swapping methods mainly rely on GAN frameworks, but recent focus has shifted to pre-trained diffusion models for their superior generation capabilities. However, training these models is resource-intensive, and the results have not yet achieved satisfactory performance levels. To address this issue, we introduce Face-Adapter, an efficient and effective adapter designed for high-precision and high-fidelity face editing for pre-trained diffusion models. We observe that both face reenactment/swapping tasks essentially involve combinations of target structure, ID and attribute. We aim to sufficiently decouple the control of these factors to achieve both tasks in one model. Specifically, our method contains: 1) A Spatial Condition Generator that provides precise landmarks and background; 2) A Plug-and-play Identity Encoder that transfers face embeddings to the text space by a transformer decoder. 3) An Attribute Controller that integrates spatial conditions and detailed attributes. Face-Adapter achieves comparable or even superior performance in terms of motion control precision, ID retention capability, and generation quality compared to fully fine-tuned face reenactment/swapping models. Additionally, Face-Adapter seamlessly integrates with various StableDiffusion models.

Yue Han, Naihong Hu

In this paper, we investigate the decay behaviors of three imaginarity-related metrics, specifically the $l_1$-norm-based imaginarity measure, imaginarity robustness, and imaginarity relative entropy, for arbitrary single-qubit pure initial states under three typical quantum channels: dephasing, generalized amplitude damping, and phase-amplitude damping. Furthermore, we extend our analysis to higher-dimensional systems by examining the decay trends of the aforementioned imaginarity metrics for several key two-qubit states under two-qubit channels. We also generalize the concept of the maximal imaginary state (originally defined for single qubits in the resource theory of imaginarity) to separable two-qubit states. In addition, we extend the definitions of imaginary power and de-imaginary power for single-qubit channels to two-qubit channels acting on separable two-qubit states. Finally, we compute the imaginary and de-imaginary powers for several common two-qubit channels.

Yue Han, Elliot Anshelevich

We study a version of the metric facility location problem (or, equivalently, variants of the committee selection problem) in which we must choose $k$ facilities in an arbitrary metric space to serve some set of clients $C$. We consider four different objectives, where each client $i\in C$ attempts to minimize either the sum or the maximum of its distance to the chosen facilities, and where the overall objective either considers the sum or the maximum of the individual client costs. Rather than optimizing a single objective at a time, we study how compatible these objectives are with each other, and show the existence of solutions which are simultaneously close-to-optimum for any pair of the above objectives. Our results show that when choosing a set of facilities or a representative committee, it is often possible to form a solution which is good for several objectives at the same time, instead of sacrificing one desideratum to achieve another.

Qilin Wang, Jiangning Zhang, Chengming Xu et al.

Facial Appearance Editing (FAE) aims to modify physical attributes, such as pose, expression and lighting, of human facial images while preserving attributes like identity and background, showing great importance in photograph. In spite of the great progress in this area, current researches generally meet three challenges: low generation fidelity, poor attribute preservation, and inefficient inference. To overcome above challenges, this paper presents DiffFAE, a one-stage and highly-efficient diffusion-based framework tailored for high-fidelity FAE. For high-fidelity query attributes transfer, we adopt Space-sensitive Physical Customization (SPC), which ensures the fidelity and generalization ability by utilizing rendering texture derived from 3D Morphable Model (3DMM). In order to preserve source attributes, we introduce the Region-responsive Semantic Composition (RSC). This module is guided to learn decoupled source-regarding features, thereby better preserving the identity and alleviating artifacts from non-facial attributes such as hair, clothes, and background. We further introduce a consistency regularization for our pipeline to enhance editing controllability by leveraging prior knowledge in the attention matrices of diffusion model. Extensive experiments demonstrate the superiority of DiffFAE over existing methods, achieving state-of-the-art performance in facial appearance editing.

Yue Han, Jiangpeng He, Mridul Gupta et al.

Image-based dietary assessment serves as an efficient and accurate solution for recording and analyzing nutrition intake using eating occasion images as input. Deep learning-based techniques are commonly used to perform image analysis such as food classification, segmentation, and portion size estimation, which rely on large amounts of food images with annotations for training. However, such data dependency poses significant barriers to real-world applications, because acquiring a substantial, diverse, and balanced set of food images can be challenging. One potential solution is to use synthetic food images for data augmentation. Although existing work has explored the use of generative adversarial networks (GAN) based structures for generation, the quality of synthetic food images still remains subpar. In addition, while diffusion-based generative models have shown promising results for general image generation tasks, the generation of food images can be challenging due to the substantial intra-class variance. In this paper, we investigate the generation of synthetic food images based on the conditional diffusion model and propose an effective clustering-based training framework, named ClusDiff, for generating high-quality and representative food images. The proposed method is evaluated on the Food-101 dataset and shows improved performance when compared with existing image generation works. We also demonstrate that the synthetic food images generated by ClusDiff can help address the severe class imbalance issue in long-tailed food classification using the VFN-LT dataset.

Xuhai Chen, Yue Han, Jiangning Zhang

In this technical report, we briefly introduce our solution for the Zero/Few-shot Track of the Visual Anomaly and Novelty Detection (VAND) 2023 Challenge. For industrial visual inspection, building a single model that can be rapidly adapted to numerous categories without or with only a few normal reference images is a promising research direction. This is primarily because of the vast variety of the product types. For the zero-shot track, we propose a solution based on the CLIP model by adding extra linear layers. These layers are used to map the image features to the joint embedding space, so that they can compare with the text features to generate the anomaly maps. Besides, when the reference images are available, we utilize multiple memory banks to store their features and compare them with the features of the test images during the testing phase. In this challenge, our method achieved first place in the zero-shot track, especially excelling in segmentation with an impressive F1 score improvement of 0.0489 over the second-ranked participant. Furthermore, in the few-shot track, we secured the fourth position overall, with our classification F1 score of 0.8687 ranking first among all participating teams.

Chao Xu, Junwei Zhu, Jiangning Zhang et al.

Recently, emotional talking face generation has received considerable attention. However, existing methods only adopt one-hot coding, image, or audio as emotion conditions, thus lacking flexible control in practical applications and failing to handle unseen emotion styles due to limited semantics. They either ignore the one-shot setting or the quality of generated faces. In this paper, we propose a more flexible and generalized framework. Specifically, we supplement the emotion style in text prompts and use an Aligned Multi-modal Emotion encoder to embed the text, image, and audio emotion modality into a unified space, which inherits rich semantic prior from CLIP. Consequently, effective multi-modal emotion space learning helps our method support arbitrary emotion modality during testing and could generalize to unseen emotion styles. Besides, an Emotion-aware Audio-to-3DMM Convertor is proposed to connect the emotion condition and the audio sequence to structural representation. A followed style-based High-fidelity Emotional Face generator is designed to generate arbitrary high-resolution realistic identities. Our texture generator hierarchically learns flow fields and animated faces in a residual manner. Extensive experiments demonstrate the flexibility and generalization of our method in emotion control and the effectiveness of high-quality face synthesis.

Jiangning Zhang, Chao Xu, Jian Li et al.

In the practical application of restoring low-resolution gray-scale images, we generally need to run three separate processes of image colorization, super-resolution, and dows-sampling operation for the target device. However, this pipeline is redundant and inefficient for the independent processes, and some inner features could have been shared. Therefore, we present an efficient paradigm to perform {S}imultaneously Image {C}olorization and {S}uper-resolution (SCS) and propose an end-to-end SCSNet to achieve this goal. The proposed method consists of two parts: colorization branch for learning color information that employs the proposed plug-and-play \emph{Pyramid Valve Cross Attention} (PVCAttn) module to aggregate feature maps between source and reference images; and super-resolution branch for integrating color and texture information to predict target images, which uses the designed \emph{Continuous Pixel Mapping} (CPM) module to predict high-resolution images at continuous magnification. Furthermore, our SCSNet supports both automatic and referential modes that is more flexible for practical application. Abundant experiments demonstrate the superiority of our method for generating authentic images over state-of-the-art methods, e.g., averagely decreasing FID by 1.8$\downarrow$ and 5.1 $\downarrow$ compared with current best scores for automatic and referential modes, respectively, while owning fewer parameters (more than $\times$2$\downarrow$) and faster running speed (more than $\times$3$\uparrow$).

Zeman Shao, Yue Han, Jiangpeng He et al.

Accurate assessment of dietary intake requires improved tools to overcome limitations of current methods including user burden and measurement error. Emerging technologies such as image-based approaches using advanced machine learning techniques coupled with widely available mobile devices present new opportunities to improve the accuracy of dietary assessment that is cost-effective, convenient and timely. However, the quality and quantity of datasets are essential for achieving good performance for automated image analysis. Building a large image dataset with high quality groundtruth annotation is a challenging problem, especially for food images as the associated nutrition information needs to be provided or verified by trained dietitians with domain knowledge. In this paper, we present the design and development of a mobile, image-based dietary assessment system to capture and analyze dietary intake, which has been deployed in both controlled-feeding and community-dwelling dietary studies. Our system is capable of collecting high quality food images in naturalistic settings and provides groundtruth annotations for developing new computational approaches.