Yiming Wu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yiming Wu has a total of 36 co-authors. Here are the top co-authors by number of shared publications:
- Ronghua Liang
College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, China
3 shared publications - Dong Xu
The University of Hong Kong, Hong Kong, Hong Kong
2 shared publications - Haoran Liang
College of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, China
2 shared publications - Qihe Pan
Zhejiang University of Technology, Hangzhou, China
2 shared publications - Sifan Long
Jilin University
2 shared publications - Xi Li
Zhejiang University, Hangzhou, China
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1015-1024
Versatile Video Representation via Feed-Forward 2D Gaussian Splatting Tokenization
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 1015-1024
Open AccessVersatile Video Representation via Feed-Forward 2D Gaussian Splatting Tokenization
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 21 (11) , 1-19
SOEDiff: Efficient Distillation for Small Object Editing
In this article, we delve into a new task known as Small Object Editing (SOE), which focuses on text-based image inpainting within a constrained, small-sized area. Despite the remarkable success have been achieved by current image inpainting approaches, their application to the SOE task generally results in failure cases such as Object Missing, Text-Image Mismatch, and Distortion . These failures stem from the limited use of small-sized objects in training datasets and the down-sampling operations employed by U-Net models, which hinders accurate generation. To overcome these challenges, we introduce a novel training-based approach, SOEDiff, aimed at enhancing the capability of baseline models like StableDiffusion in editing small-sized objects while minimizing training costs. Specifically, our method involves two key components: SO-LoRA , which efficiently fine-tunes low-rank matrices, and Cross-scale score distillation , which leverages high-resolution predictions from the pre-trained teacher diffusion model. Our method presents significant improvements on the test dataset collected from MSCOCO and OpenImage, validating the effectiveness of our proposed method in SOE. In particular, when comparing SOEDiff with SD-I model on the OpenImage-small-val dataset, we observe a 0.99 improvement in CLIP-Score and a reduction of 2.87 in FID.
- 10 November 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 21 (11) , 1-19
Open AccessSOEDiff: Efficient Distillation for Small Object Editing
In this article, we delve into a new task known as Small Object Editing (SOE), which focuses on text-based image inpainting within a constrained, small-sized area. Despite the remarkable success have been achieved by current image inpainting approaches, their application to the SOE task generally results in failure cases such as Object Missing, Text-Image Mismatch, and Distortion . These failures stem from the limited use of small-sized objects in training datasets and the down-sampling operations employed by U-Net models, which hinders accurate generation. To overcome these challenges, we introduce a novel training-based approach, SOEDiff, aimed at enhancing the capability of baseline models like StableDiffusion in editing small-sized objects while minimizing training costs. Specifically, our method involves two key components: SO-LoRA , which efficiently fine-tunes low-rank matrices, and Cross-scale score distillation , which leverages high-resolution predictions from the pre-trained teacher diffusion model. Our method presents significant improvements on the test dataset collected from MSCOCO and OpenImage, validating the effectiveness of our proposed method in SOE. In particular, when comparing SOEDiff with SD-I model on the OpenImage-small-val dataset, we observe a 0.99 improvement in CLIP-Score and a reduction of 2.87 in FID.
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9714-9723
Individual Content and Motion Dynamics Preserved Pruning for Video Diffusion Models
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9714-9723
Open AccessIndividual Content and Motion Dynamics Preserved Pruning for Video Diffusion Models
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 3257-3265
Towards Small Object Editing: A Benchmark Dataset and A Training-Free Approach
A plethora of text-guided image editing methods has recently been developed by leveraging the impressive capabilities of large-scale diffusion-based generative models especially Stable Diffusion. Despite the success of diffusion models in producing high-quality images, their application to small object generation has been limited due to difficulties in aligning cross-modal attention maps between text and these objects. Our approach offers a training-free method that significantly mitigates this alignment issue with local and global attention guidance, enhancing the model's ability to accurately render small objects in accordance with textual descriptions. We detail the methodology in our approach, emphasizing its divergence from traditional generation techniques and highlighting its advantages. What's more important is that we also provide SOEBench (Small Object Editing), a standardized benchmark for quantitatively evaluating text-based small object generation collected from MSCOCO[22] and OpenImage[18]. Preliminary results demonstrate the effectiveness of our method, showing marked improvements in the fidelity and accuracy of small object generation compared to existing models. This advancement not only contributes to the field of AI and computer vision but also opens up new possibilities for applications in various industries where precise image generation is critical.We will release our dataset on our project page: https://soebench.github.io/
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 3257-3265
Open AccessTowards Small Object Editing: A Benchmark Dataset and A Training-Free Approach
A plethora of text-guided image editing methods has recently been developed by leveraging the impressive capabilities of large-scale diffusion-based generative models especially Stable Diffusion. Despite the success of diffusion models in producing high-quality images, their application to small object generation has been limited due to difficulties in aligning cross-modal attention maps between text and these objects. Our approach offers a training-free method that significantly mitigates this alignment issue with local and global attention guidance, enhancing the model's ability to accurately render small objects in accordance with textual descriptions. We detail the methodology in our approach, emphasizing its divergence from traditional generation techniques and highlighting its advantages. What's more important is that we also provide SOEBench (Small Object Editing), a standardized benchmark for quantitatively evaluating text-based small object generation collected from MSCOCO[22] and OpenImage[18]. Preliminary results demonstrate the effectiveness of our method, showing marked improvements in the fidelity and accuracy of small object generation compared to existing models. This advancement not only contributes to the field of AI and computer vision but also opens up new possibilities for applications in various industries where precise image generation is critical.We will release our dataset on our project page: https://soebench.github.io/
- Open Access
- 30 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1270-1274
RE-IDVIS: Person Re-Identification System based on Interactive Visualization
Person re-identification (Re-ID) is a crucial technique in security systems. It enables the accurate identification and tracking of specific individuals from multiple camera perspectives. Despite significant progress in existing person Re-ID methods, they often struggle to achieve expected results in person retrieval in real-world application scenarios. One of the main challenges is that these models usually require manual adjustments to adapt to specific cases, and users often find it difficult to communicate their insights and requirements to the models. Additionally, model results include thousands of candidate images, making it challenging for users to browse and retrieve them efficiently. To address these issues, we propose a visual analysis system, RE-IDVIS, to bridge the communication gap between humans and machine learning models. This system integrates an unsupervised person Re-ID model, a semi-supervised human-in-the-loop re-ranking algorithm, and a series of visual designs and interactions to guide users in real-world person retrieval.
- 30 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1270-1274
Open AccessRE-IDVIS: Person Re-Identification System based on Interactive Visualization
Person re-identification (Re-ID) is a crucial technique in security systems. It enables the accurate identification and tracking of specific individuals from multiple camera perspectives. Despite significant progress in existing person Re-ID methods, they often struggle to achieve expected results in person retrieval in real-world application scenarios. One of the main challenges is that these models usually require manual adjustments to adapt to specific cases, and users often find it difficult to communicate their insights and requirements to the models. Additionally, model results include thousands of candidate images, making it challenging for users to browse and retrieve them efficiently. To address these issues, we propose a visual analysis system, RE-IDVIS, to bridge the communication gap between humans and machine learning models. This system integrates an unsupervised person Re-ID model, a semi-supervised human-in-the-loop re-ranking algorithm, and a series of visual designs and interactions to guide users in real-world person retrieval.
- Open Access
- 29 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 49-57
Dynamic Network for Language-based Fashion Retrieval
Language-based fashion image retrieval, as a kind of composed image retrieval, presents a substantial challenge in the domain of multi-modal retrieval. This task aims to retrieve the target fashion item in the gallery given a reference image and a modification text. Existing approaches primarily concentrate on developing a static multi-modal fusion module to learn the combined semantics of the reference image and modification text. Despite their commendable advancements, these approaches are still limited by a deficiency in flexibility, which is attributed to the application of a singluar fusion module across diverse input queries. In contrast to static fusion methods, we propose a novel method termed Dynamic Fusion Network (DFN) to compose the multi-granularity features dynamically by considering the consistency of routing path and modality-specific information simultaneously. In specific, our proposed method is consisted of two modules: (1) Dynamic Network. The dynamic network enables a flexible combination of the different operation modules, providing multi-granularity modality interaction for each reference image and modifier text. (2) Modality Specific Routers (MSR). The MSR generates precise routing decisions based on the distinct semantics and distributions of each reference image and modifier text. Extensive experiments on three benchmarks,\textiti.e., FashionIQ, Shoes and Fashion200K, demonstrate the effectiveness of our proposed model compared with the existing methods.
- 29 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 49-57
Open AccessDynamic Network for Language-based Fashion Retrieval
Language-based fashion image retrieval, as a kind of composed image retrieval, presents a substantial challenge in the domain of multi-modal retrieval. This task aims to retrieve the target fashion item in the gallery given a reference image and a modification text. Existing approaches primarily concentrate on developing a static multi-modal fusion module to learn the combined semantics of the reference image and modification text. Despite their commendable advancements, these approaches are still limited by a deficiency in flexibility, which is attributed to the application of a singluar fusion module across diverse input queries. In contrast to static fusion methods, we propose a novel method termed Dynamic Fusion Network (DFN) to compose the multi-granularity features dynamically by considering the consistency of routing path and modality-specific information simultaneously. In specific, our proposed method is consisted of two modules: (1) Dynamic Network. The dynamic network enables a flexible combination of the different operation modules, providing multi-granularity modality interaction for each reference image and modifier text. (2) Modality Specific Routers (MSR). The MSR generates precise routing decisions based on the distinct semantics and distributions of each reference image and modifier text. Extensive experiments on three benchmarks,\textiti.e., FashionIQ, Shoes and Fashion200K, demonstrate the effectiveness of our proposed model compared with the existing methods.
- Open Access
- 15 March 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 19 (4) , 1-20
D 3 T-GAN: Data-Dependent Domain Transfer GANs for Image Generation with Limited Data
As an important and challenging problem, image generation with limited data aims at generating realistic images through training a GAN model given few samples. A typical solution is to transfer a well-trained GAN model from a data-rich source domain to the data-deficient target domain. In this paper, we propose a novel self-supervised transfer scheme termed D 3 T-GAN, addressing the cross-domain GANs transfer in limited image generation. Specifically, we design two individual strategies to transfer knowledge between generators and discriminators, respectively. To transfer knowledge between generators, we conduct a data-dependent transformation, which projects target samples into the latent space of source generator and reconstructs them back. Then, we perform knowledge transfer from transformed samples to generated samples. To transfer knowledge between discriminators, we design a multi-level discriminant knowledge distillation from the source discriminator to the target discriminator on both the real and fake samples. Extensive experiments show that our method improves the quality of generated images and achieves the state-of-the-art FID scores on commonly used datasets.
- 15 March 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 19 (4) , 1-20
Open AccessD 3 T-GAN: Data-Dependent Domain Transfer GANs for Image Generation with Limited Data
As an important and challenging problem, image generation with limited data aims at generating realistic images through training a GAN model given few samples. A typical solution is to transfer a well-trained GAN model from a data-rich source domain to the data-deficient target domain. In this paper, we propose a novel self-supervised transfer scheme termed D 3 T-GAN, addressing the cross-domain GANs transfer in limited image generation. Specifically, we design two individual strategies to transfer knowledge between generators and discriminators, respectively. To transfer knowledge between generators, we conduct a data-dependent transformation, which projects target samples into the latent space of source generator and reconstructs them back. Then, we perform knowledge transfer from transformed samples to generated samples. To transfer knowledge between discriminators, we design a multi-level discriminant knowledge distillation from the source discriminator to the target discriminator on both the real and fake samples. Extensive experiments show that our method improves the quality of generated images and achieves the state-of-the-art FID scores on commonly used datasets.
- 1 January 2023
- Published by Medknow in Journal of Radiological Science
- Vol. 48 (1) , 7
Comparison of the artificial intelligence in characterizing pulmonary nodule on low-dose computed tomograms using different slice thicknesses
- 1 January 2023
- Published by Medknow in Journal of Radiological Science
- Vol. 48 (1) , 7
Comparison of the artificial intelligence in characterizing pulmonary nodule on low-dose computed tomograms using different slice thicknesses
- 23 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 30 (10577149) , 8658-8670
F³A-GAN: Facial Flow for Face Animation With Generative Adversarial Networks
Formulated as a conditional generation problem, face animation aims at synthesizing continuous face images from a single source image driven by a set of conditional face motion. Previous works mainly model the face motion as conditions with 1D or 2D representation ( e.g., action units, emotion codes, landmark ), which often leads to low-quality results in some complicated scenarios such as continuous generation and large-pose transformation. To tackle this problem, the conditions are supposed to meet two requirements, i.e., motion information preserving and geometric continuity. To this end, we propose a novel representation based on a 3D geometric flow, termed facial flow, to represent the natural motion of the human face at any pose. Compared with other previous conditions, the proposed facial flow well controls the continuous changes to the face. After that, in order to utilize the facial flow for face editing, we build a synthesis framework generating continuous images with conditional facial flows. To fully take advantage of the motion information of facial flows, a hierarchical conditional framework is designed to combine the extracted multi-scale appearance features from images and motion features from flows in a hierarchical manner. The framework then decodes multiple fused features back to images progressively. Experimental results demonstrate the effectiveness of our method compared to other state-of-the-art methods.
- 23 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 30 (10577149) , 8658-8670
F³A-GAN: Facial Flow for Face Animation With Generative Adversarial Networks
Formulated as a conditional generation problem, face animation aims at synthesizing continuous face images from a single source image driven by a set of conditional face motion. Previous works mainly model the face motion as conditions with 1D or 2D representation ( e.g., action units, emotion codes, landmark ), which often leads to low-quality results in some complicated scenarios such as continuous generation and large-pose transformation. To tackle this problem, the conditions are supposed to meet two requirements, i.e., motion information preserving and geometric continuity. To this end, we propose a novel representation based on a 3D geometric flow, termed facial flow, to represent the natural motion of the human face at any pose. Compared with other previous conditions, the proposed facial flow well controls the continuous changes to the face. After that, in order to utilize the facial flow for face editing, we build a synthesis framework generating continuous images with conditional facial flows. To fully take advantage of the motion information of facial flows, a hierarchical conditional framework is designed to combine the extracted multi-scale appearance features from images and motion features from flows in a hierarchical manner. The framework then decodes multiple fused features back to images progressively. Experimental results demonstrate the effectiveness of our method compared to other state-of-the-art methods.
- Open Access
- 8 July 2021
- Published by Association for Computing Machinery (ACM)
- p. 32-37
Design and Analysis of Blood Donation Model Based on Blockchain and KNN
With the development of blood donation industry in China, the blood donation model has changed. Outdated blood donation management software can no longer meet the needs of the new era, so it is imperative to design and implement a blood donation platform based on the combination of blockchain and KNN(K-Nearest Neighbor).This system uses blockchain to store the key information of blood donation and Hash file information, uses IPFS to store the identity image and signature file of the real blood donor, and uses the KNN algorithm to match the nearest blood supply center with a specific blood type, in order to deploy timely when the hospital is ischemic and reduce the mortality caused by insufficient blood in the blood bank. This project combines traditional blood donation with blockchain and KNN. It can not only ensure the privacy and safety of blood donors, and trace the blood source, but also can match the nearest blood bank to call the hospital for ischemia, which will contribute to the public welfare undertakings of blood donation. I hope this project can provide a reference for the future blood donation business.
- 8 July 2021
- Published by Association for Computing Machinery (ACM)
- p. 32-37
Open AccessDesign and Analysis of Blood Donation Model Based on Blockchain and KNN
With the development of blood donation industry in China, the blood donation model has changed. Outdated blood donation management software can no longer meet the needs of the new era, so it is imperative to design and implement a blood donation platform based on the combination of blockchain and KNN(K-Nearest Neighbor).This system uses blockchain to store the key information of blood donation and Hash file information, uses IPFS to store the identity image and signature file of the real blood donor, and uses the KNN algorithm to match the nearest blood supply center with a specific blood type, in order to deploy timely when the hospital is ischemic and reduce the mortality caused by insufficient blood in the blood bank. This project combines traditional blood donation with blockchain and KNN. It can not only ensure the privacy and safety of blood donors, and trace the blood source, but also can match the nearest blood bank to call the hospital for ischemia, which will contribute to the public welfare undertakings of blood donation. I hope this project can provide a reference for the future blood donation business.