Shan Du

Wenqi Marshall Guo, Qingyun Qian, Khalad Hasan et al.

Over-aligning image generation models to a generalized aesthetic preference conflicts with user intent, particularly when "anti-aesthetic" outputs are requested for artistic or critical purposes. This adherence prioritizes developer-centered values, compromising user autonomy and aesthetic pluralism. We test this bias by constructing a wide-spectrum aesthetics dataset and evaluating state-of-the-art generation and reward models. This position paper finds that aesthetic-aligned generation models frequently default to conventionally beautiful outputs, failing to respect instructions for low-quality or negative imagery. Crucially, reward models penalize anti-aesthetic images even when they perfectly match the explicit user prompt. We confirm this systemic bias through image-to-image editing and evaluation against real abstract artworks. Our code, fine-tuned models, and datasets are available on our meta-expression intentionally anti-aesthetics webpage: https://weathon.github.io/icml2026_position/.

Xiang Wang, Yimin Yang, Zhichang Guo et al.

Deep Convolutional Neural Networks (DCNNs) have exhibited impressive performance on image super-resolution tasks. However, these deep learning-based super-resolution methods perform poorly in real-world super-resolution tasks, where the paired high-resolution and low-resolution images are unavailable and the low-resolution images are degraded by complicated and unknown kernels. To break these limitations, we propose the Unsupervised Bi-directional Cycle Domain Transfer Learning-based Generative Adversarial Network (UBCDTL-GAN), which consists of an Unsupervised Bi-directional Cycle Domain Transfer Network (UBCDTN) and the Semantic Encoder guided Super Resolution Network (SESRN). First, the UBCDTN is able to produce an approximated real-like LR image through transferring the LR image from an artificially degraded domain to the real-world LR image domain. Second, the SESRN has the ability to super-resolve the approximated real-like LR image to a photo-realistic HR image. Extensive experiments on unpaired real-world image benchmark datasets demonstrate that the proposed method achieves superior performance compared to state-of-the-art methods.

Xiang Wang, Yimin Yang, Qixiang Pang et al.

Face super-resolution is a domain-specific image super-resolution, which aims to generate High-Resolution (HR) face images from their Low-Resolution (LR) counterparts. In this paper, we propose a novel face super-resolution method, namely Semantic Encoder guided Generative Adversarial Face Ultra-Resolution Network (SEGA-FURN) to ultra-resolve an unaligned tiny LR face image to its HR counterpart with multiple ultra-upscaling factors (e.g., 4x and 8x). The proposed network is composed of a novel semantic encoder that has the ability to capture the embedded semantics to guide adversarial learning and a novel generator that uses a hierarchical architecture named Residual in Internal Dense Block (RIDB). Moreover, we propose a joint discriminator which discriminates both image data and embedded semantics. The joint discriminator learns the joint probability distribution of the image space and latent space. We also use a Relativistic average Least Squares loss (RaLS) as the adversarial loss to alleviate the gradient vanishing problem and enhance the stability of the training procedure. Extensive experiments on large face datasets have proved that the proposed method can achieve superior super-resolution results and significantly outperform other state-of-the-art methods in both qualitative and quantitative comparisons.

Xinnan Ding, Shan Du, Yu Zhang et al.

The critical goal of gait recognition is to acquire the inter-frame walking habit representation from the gait sequences. The relations between frames, however, have not received adequate attention in comparison to the intra-frame features. In this paper, motivated by optical flow, the bilateral motion-oriented features are proposed, which can allow the classic convolutional structure to have the capability to directly portray gait movement patterns at the feature level. Based on such features, we develop a set of multi-scale temporal representations that force the motion context to be richly described at various levels of temporal resolution. Furthermore, a correction block is devised to eliminate the segmentation noise of silhouettes for getting more precise gait information. Subsequently, the temporal feature set and the spatial features are combined to comprehensively characterize gait processes. Extensive experiments are conducted on CASIA-B and OU-MVLP datasets, and the results achieve an outstanding identification performance, which has demonstrated the effectiveness of the proposed approach.

Dan Wang, Haiyan Sun, Shan Du et al.

Image super-resolution (SR) aims to reconstruct high resolution images with both high perceptual quality and low distortion, but is fundamentally limited by the perception-distortion trade-off. GAN-based SR methods reduce distortion but still struggle with realistic fine-grained textures, whereas diffusion-based approaches synthesize rich details but often deviate from the input, hallucinating structures and degrading fidelity. This tension raises a key challenge: how to exploit the powerful generative priors of diffusion models without sacrificing fidelity. To address this, we propose SpaSemSR, a spatial-semantic guided diffusion framework with two complementary guidances. First, spatial-grounded textual guidance integrates object-level spatial cues with semantic prompts, aligning textual and visual structures to reduce distortion. Second, semantic-enhanced visual guidance with a multi-encoder design and semantic degradation constraints unifies multimodal semantic priors, improving perceptual realism under severe degradations. These complementary guidances are adaptively fused into the diffusion process via spatial-semantic attention, suppressing distortion and hallucination while retaining the strengths of diffusion models. Extensive experiments on multiple benchmarks show that SpaSemSR achieves a superior perception-distortion balance, producing both realistic and faithful restorations.

Wenqi Guo, Yiyang Du, Shan Du

Gas leakage poses a significant hazard that requires prevention. Traditionally, human inspection has been used for detection, a slow and labour-intensive process. Recent research has applied machine learning techniques to this problem, yet there remains a shortage of high-quality, publicly available datasets. This paper introduces a synthetic dataset, SimGas, featuring diverse backgrounds, interfering foreground objects, diverse leak locations, and precise segmentation ground truth. We propose a zero-shot method that combines background subtraction, zero-shot object detection, filtering, and segmentation to leverage this dataset. Experimental results indicate that our approach significantly outperforms baseline methods based solely on background subtraction and zero-shot object detection with segmentation, reaching an IoU of 69%. We also present an analysis of various prompt configurations and threshold settings to provide deeper insights into the performance of our method. Finally, we qualitatively (because of the lack of ground truth) tested our performance on GasVid and reached decent results on the real-world dataset. The dataset, code, and full qualitative results are available at https://github.com/weathon/Lang-Gas.

Peizhi Yan, Rabab Ward, Qiang Tang et al.

Recent advancements in 3D Gaussian Splatting (3DGS) have unlocked significant potential for modeling 3D head avatars, providing greater flexibility than mesh-based methods and more efficient rendering compared to NeRF-based approaches. Despite these advancements, the creation of controllable 3DGS-based head avatars remains time-intensive, often requiring tens of minutes to hours. To expedite this process, we here introduce the "Gaussian Deja-vu" framework, which first obtains a generalized model of the head avatar and then personalizes the result. The generalized model is trained on large 2D (synthetic and real) image datasets. This model provides a well-initialized 3D Gaussian head that is further refined using a monocular video to achieve the personalized head avatar. For personalizing, we propose learnable expression-aware rectification blendmaps to correct the initial 3D Gaussians, ensuring rapid convergence without the reliance on neural networks. Experiments demonstrate that the proposed method meets its objectives. It outperforms state-of-the-art 3D Gaussian head avatars in terms of photorealistic quality as well as reduces training time consumption to at least a quarter of the existing methods, producing the avatar in minutes.

Xinlong Zhao, Qixiang Pang, Shan Du

Gas leaks pose serious threats to human health and contribute significantly to atmospheric pollution, drawing increasing public concern. However, the lack of effective detection methods hampers timely and accurate identification of gas leaks. While some vision-based techniques leverage infrared videos for leak detection, the blurry and non-rigid nature of gas clouds often limits their effectiveness. To address these challenges, we propose a novel framework called Joint Vision-Language Gas leak Segmentation (JVLGS), which integrates the complementary strengths of visual and textual modalities to enhance gas leak representation and segmentation. Recognizing that gas leaks are sporadic and many video frames may contain no leak at all, our method incorporates a post-processing step to reduce false positives caused by noise and non-target objects, an issue that affects many existing approaches. Extensive experiments conducted across diverse scenarios show that JVLGS significantly outperforms state-of-the-art gas leak segmentation methods. We evaluate our model under both supervised and few-shot learning settings, and it consistently achieves strong performance in both, whereas competing methods tend to perform well in only one setting or poorly in both. Code available at: https://github.com/GeekEagle/JVLGS

Wenqi Guo, Shan Du

We introduce Value Sign Flip (VSF), a simple and efficient method for incorporating negative prompt guidance in few-step diffusion and flow-matching image generation models. Unlike existing approaches such as classifier-free guidance (CFG), NASA, and NAG, VSF dynamically suppresses undesired content by flipping the sign of attention values from negative prompts. Our method requires only small computational overhead and integrates effectively with MMDiT-style architectures such as Stable Diffusion 3.5 Turbo, as well as cross-attention-based models like Wan. We validate VSF on challenging datasets with complex prompt pairs and demonstrate superior performance in both static image and video generation tasks. Experimental results show that VSF significantly improves negative prompt adherence compared to prior methods in few-step models, and even CFG in non-few-step models, while maintaining competitive image quality. Code and ComfyUI node are available in https://github.com/weathon/VSF/tree/main.

Wenqi Guo, Mohamed Shehata, Shan Du

Camouflaged object segmentation presents unique challenges compared to traditional segmentation tasks, primarily due to the high similarity in patterns and colors between camouflaged objects and their backgrounds. Effective solutions to this problem have significant implications in critical areas such as pest control, defect detection, and lesion segmentation in medical imaging. Prior research has predominantly emphasized supervised or unsupervised pre-training methods, leaving zero-shot approaches significantly underdeveloped. Existing zero-shot techniques commonly utilize the Segment Anything Model (SAM) in automatic mode or rely on vision-language models to generate cues for segmentation; however, their performances remain unsatisfactory, due to the similarity of the camouflaged object and the background. This work studies how to avoid training by integrating large pre-trained models like SAM-2 and Owl-v2 with temporal information into a modular pipeline. Evaluated on the MoCA-Mask dataset, our approach achieves outstanding performance improvements, significantly outperforming existing zero-shot methods by raising the F-measure ($F_β^w$) from 0.296 to 0.628. Our approach also surpasses supervised methods, increasing the F-measure from 0.476 to 0.628. Additionally, evaluation on the MoCA-Filter dataset demonstrates an increase in the success rate from 0.628 to 0.697 when compared with FlowSAM, a supervised transfer method. A thorough ablation study further validates the individual contributions of each component. Besides our main contributions, we also highlight inconsistencies in previous work regarding metrics and settings. Code can be found in https://github.com/weathon/vcos.

Peizhi Yan, Rabab K. Ward, Dan Wang et al.

For 3D face modeling, the recently developed 3D-aware neural rendering methods are able to render photorealistic face images with arbitrary viewing directions. The training of the parametric controllable 3D-aware face models, however, still relies on a large-scale dataset that is lab-collected. To address this issue, this paper introduces "StyleMorpheus", the first style-based neural 3D Morphable Face Model (3DMM) that is trained on in-the-wild images. It inherits 3DMM's disentangled controllability (over face identity, expression, and appearance) but without the need for accurately reconstructed explicit 3D shapes. StyleMorpheus employs an auto-encoder structure. The encoder aims at learning a representative disentangled parametric code space and the decoder improves the disentanglement using shape and appearance-related style codes in the different sub-modules of the network. Furthermore, we fine-tune the decoder through style-based generative adversarial learning to achieve photorealistic 3D rendering quality. The proposed style-based design enables StyleMorpheus to achieve state-of-the-art 3D-aware face reconstruction results, while also allowing disentangled control of the reconstructed face. Our model achieves real-time rendering speed, allowing its use in virtual reality applications. We also demonstrate the capability of the proposed style-based design in face editing applications such as style mixing and color editing. Project homepage: https://github.com/ubc-3d-vision-lab/StyleMorpheus.

Xiangrui Liu, Yuanyuan Zhang, Qianyu Shang et al.

Foundation models, first introduced in 2021, refer to large-scale pretrained models (e.g., large language models (LLMs) and vision-language models (VLMs)) that learn from extensive unlabeled datasets through unsupervised methods, enabling them to excel in diverse downstream tasks. These models, like GPT, can be adapted to various applications such as question answering and visual understanding, outperforming task-specific AI models and earning their name due to broad applicability across fields. The development of biomedical foundation models marks a significant milestone in the use of artificial intelligence (AI) to understand complex biological phenomena and advance medical research and practice. This survey explores the potential of foundation models in diverse domains within biomedical fields, including computational biology, drug discovery and development, clinical informatics, medical imaging, and public health. The purpose of this survey is to inspire ongoing research in the application of foundation models to health science.

Xinlong Zhao, Shan Du

Gas leaks pose significant risks to human health and the environment. Despite long-standing concerns, there are limited methods that can efficiently and accurately detect and segment leaks due to their concealed appearance and random shapes. In this paper, we propose a Fine-grained Spatial-Temporal Perception (FGSTP) algorithm for gas leak segmentation. FGSTP captures critical motion clues across frames and integrates them with refined object features in an end-to-end network. Specifically, we first construct a correlation volume to capture motion information between consecutive frames. Then, the fine-grained perception progressively refines the object-level features using previous outputs. Finally, a decoder is employed to optimize boundary segmentation. Because there is no highly precise labeled dataset for gas leak segmentation, we manually label a gas leak video dataset, GasVid. Experimental results on GasVid demonstrate that our model excels in segmenting non-rigid objects such as gas leaks, generating the most accurate mask compared to other state-of-the-art (SOTA) models.

Wenqi Guo, Shan Du

Transformation-based privacy-preserving face recognition (PPFR) aims to verify identities while hiding facial data from attackers and malicious service providers. Existing evaluations mostly treat privacy as resistance to pixel-level reconstruction, measured by PSNR and SSIM. We show that this reconstruction-centric view fails. We present FaceLinkGen, an identity extraction attack that performs linkage/matching and face regeneration directly from protected templates without recovering original pixels. On three recent PPFR systems, FaceLinkGen reaches over 98.5\% matching accuracy and above 96\% regeneration success, and still exceeds 92\% matching and 94\% regeneration in a near zero knowledge setting. These results expose a structural gap between pixel distortion metrics, which are widely used in PPFR evaluation, and real privacy. We show that visual obfuscation leaves identity information broadly exposed to both external intruders and untrusted service providers.

Peizhi Yan, Rabab Ward, Qiang Tang et al.

3D Gaussian Splatting (3DGS) has enabled photorealistic and real-time rendering of 3D head avatars. Existing 3DGS-based avatars typically rely on tens of thousands of 3D Gaussian points (Gaussians), with the number of Gaussians fixed after training. However, many practical applications require adjustable levels of detail (LOD) to balance rendering efficiency and visual quality. In this work, we propose "ArchitectHead", the first framework for creating 3D Gaussian head avatars that support continuous control over LOD. Our key idea is to parameterize the Gaussians in a 2D UV feature space and propose a UV feature field composed of multi-level learnable feature maps to encode their latent features. A lightweight neural network-based decoder then transforms these latent features into 3D Gaussian attributes for rendering. ArchitectHead controls the number of Gaussians by dynamically resampling feature maps from the UV feature field at the desired resolutions. This method enables efficient and continuous control of LOD without retraining. Experimental results show that ArchitectHead achieves state-of-the-art (SOTA) quality in self and cross-identity reenactment tasks at the highest LOD, while maintaining near SOTA performance at lower LODs. At the lowest LOD, our method uses only 6.2\% of the Gaussians while the quality degrades moderately (L1 Loss +7.9\%, PSNR --0.97\%, SSIM --0.6\%, LPIPS Loss +24.1\%), and the rendering speed nearly doubles.