Author name cluster

Mingyang Wu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

2 author rows

NeurIPS Conference 2025 Conference Paper

4KAgent: Agentic Any Image to 4K Super-Resolution

Yushen Zuo
Qi Zheng
Mingyang Wu
Xinrui Jiang
Renjie Li
Jian Wang
Yide Zhang
Gengchen Mai

We present 4KAgent, a unified agentic super-resolution generalist system designed to universally upscale any image to 4K resolution (and even higher, if applied iteratively). Our system can transform images from extremely low resolutions with severe degradations, for example, highly distorted inputs at $256\times 256$, into crystal-clear, photorealistic 4K outputs. 4KAgent comprises three core components: (1) Profiling, a module that customizes the 4KAgent pipeline based on bespoke use cases; (2) A Perception Agent, which leverages vision-language models alongside image quality assessment experts to analyze the input image and make a tailored restoration plan; and (3) A Restoration Agent, which executes the plan, following a recursive execution-reflection paradigm, guided by a quality-driven mixture-of-experts policy to select the optimal output for each step. Additionally, 4KAgent embeds a specialized face restoration pipeline, significantly enhancing facial details in portrait and selfie photos. We rigorously evaluate our 4KAgent across 11 distinct task categories encompassing a total of 26 diverse benchmarks, setting new state-of-the-art on a broad spectrum of imaging domains. Our evaluations cover natural images, portrait photos, AI-generated content, satellite imagery, fluorescence microscopy, and medical imaging like fundoscopy, ultrasound, and X-ray, demonstrating superior performance in terms of both perceptual (e. g. , NIQE, MUSIQ) and fidelity (e. g. , PSNR) metrics. By establishing a novel agentic paradigm for low-level vision tasks, we aim to catalyze broader interest and innovation within vision-centric autonomous agents across diverse research communities. We release all the code, models, and results at: https: //4kagent. github. io.

PDF Details

EAAI Journal 2025 Journal Article

Cross-domain facial expression recognition: Bi-Directional Fusion of Active and Stable Information

Yanan Zhu
Jiaqiu Ai
Weibao Xue
Mingyang Wu
Sen Yang
Wei Jia
Min Hu

Facial expression recognition (FER) algorithms often encounter obstacles in cross-domain scenarios, attributed to variations in collection conditions such as lighting, weather, age, gender, and skin color of subjects. Unlike existing approaches that primarily focus on extracting globally invariant features and aligning domain distributions, we propose a novel framework that fundamentally shifts the approach to cross-domain FER. Our proposed algorithm, termed Bi-Directional Fusion of Active and Stable Information (FER-DAS), uniquely combines three innovative components: the Active Assessment Strategy (AAS), Cross-Domain Dynamic Class Threshold (CD-DCT), and Weighted Cross-Domain Alignment (WCDA). The AAS component selectively identifies and enhances active samples in the target domain, providing precise annotations for improved model robustness. Samples with the highest uncertainty are deemed active, indicating low prediction confidence and high informational value for model training. These are then filtered using a predefined threshold to ensure only the most informative samples are included in training iterations. In contrast to conventional static threshold techniques, our dynamic class threshold strategy (CD-DCT) adaptively filters stable samples across domains, thereby ensuring that only the most reliable information is utilized in training. The WCDA strategy further refines this process by dynamically assessing and weighting the contribution of target domain samples to class centers, effectively mitigating domain distribution discrepancies. Extensive experiments on multiple benchmark datasets confirm that FER-DAS sets a new standard in cross-domain FER, consistently outperforming existing state-of-the-art methods.

Details DOI

ICML Conference 2025 Conference Paper

Preserving AUC Fairness in Learning with Noisy Protected Groups

Mingyang Wu
Li Lin
Wenbin Zhang 0002
Xin Wang 0045
Zhenhuan Yang
Shu Hu 0001

The Area Under the ROC Curve (AUC) is a key metric for classification, especially under class imbalance, with growing research focus on optimizing AUC over accuracy in applications like medical image analysis and deepfake detection. This leads to fairness in AUC optimization becoming crucial as biases can impact protected groups. While various fairness mitigation techniques exist, fairness considerations in AUC optimization remain in their early stages, with most research focusing on improving AUC fairness under the assumption of clean protected groups. However, these studies often overlook the impact of noisy protected groups, leading to fairness violations in practice. To address this, we propose the first robust AUC fairness approach under noisy protected groups with fairness theoretical guarantees using distributionally robust optimization. Extensive experiments on tabular and image datasets show that our method outperforms state-of-the-art approaches in preserving AUC fairness. The code is in https: //github. com/Purdue-M2/AUC_Fairness_with_Noisy_Groups.

Details