Xiaojie Wang

Xiaojie Wang

Auto-generated profile, based on publicly available information

Xiaojie Wang

Beijing University of Posts and Telecommunications, Beijing, China

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 October 2024
Verified
Conference Paper
Open access
PZPengfei Zhou
Pengfei Zhou
FFFangxiang Feng
Fangxiang Feng
GLGuang Liu
Guang Liu
Ruifan LiRuifan Li
Ruifan Li
XWXiaojie Wang
Xiaojie Wang
Latent diffusion model has demonstrated impressive efficacy in image generation and editing tasks. Recently, it has also promoted the advancement of image harmonization. However, methods involving latent diffusion model all face a common challenge: the severe image distortion introduced by the VAE component, while image harmonization is a low-level image processing task that relies on pixel-level evaluation metrics. In this paper, we propose Harmony-VAE, leveraging the input of the harmonization task itself to enhance the quality of decoded images. The input involving composite image contains the precise pixel level information, which can complement the correct foreground appearance and color information contained in denoised latents. Meanwhile, the inherent generative nature of diffusion models makes it naturally adapt to inverse image harmonization, i.e. generating synthetic composite images based on real images and foreground masks. We train an inverse harmonization diffusion model to perform data augmentation on two subsets of iHarmony4 and construct a new human harmonization dataset with prominent foreground objects. Extensive experiments demonstrate the effectiveness of our proposed Harmony-VAE and inverse harmonization model. Code and pretrained models are available at https://github.com/nicecv/DiffHarmony.

28 October 2024
Verified
Conference Paper
Open access
HWHanzi Wang
Hanzi Wang
JRJiamin Ren
Jiamin Ren
YDYifeng Ding
Yifeng Ding
LRLei Ren
Lei Ren
HJHuixing Jiang
Huixing Jiang
XWXiaojie Wang
Xiaojie Wang
Multimodal Large Language Models (MLLMs) have showcased remarkable advances in handling various vision-language tasks. These models typically consist of a Large Language Model (LLM), a vision encoder and a connector structure, which is used to bridge the modality gap between vision and language. It is challenging for the connector to filter the right visual information for LLM according to the task in hand. Most of previous connectors, such as light-weight projection and Q-former, treat visual information for diverse tasks uniformly, therefore lacking task-specific visual information extraction capabilities. To address the issue, this paper proposes Q-MoE, a query-based connector with Mixture-of-Experts (MoE) to extract task-specific information with text-driven routing. Furthermore, an optimal path based training strategy is proposed to find an optimal expert combination. Extensive experiments on two popular open-source LLMs and several different visual-language tasks demonstrate the effectiveness of the Q-MoE connecter.

12 November 2025
Verified
Conference Paper
Open access
FZFangkun Zhao
Fangkun Zhao
SCSheng Chen
Sheng Chen
JHJinyu Hou
Jinyu Hou
HSHaochen Sun
Haochen Sun
QZQiang Zhang
Qiang Zhang
XWXiaojie Wang
Xiaojie Wang