Ju Yong Chang
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
Ju Yong Chang has a total of 60 co-authors. Here are the top co-authors by number of shared publications:
Kyoung Mu Lee
11 shared publicationsGyeongsik Moon
5 shared publicationsSang Uk Lee
Department of Electrical Engineering and Computer Science, ASRI, Seoul National University, Seoul 151-742, Republic of Korea
5 shared publications- Kyoung Mu Lee
Seoul National University
4 shared publications - Sungho Chun
Kwangwoon University
3 shared publications - Do Yeop Kim
Department of Electronics and Communications Engineering, Kwangwoon University, Seoul, South Korea
2 shared publications
Latest Publications
- Open Access
- 31 March 2026
- Published by The Korean Institute of Broadcast and Media Engineers in Journal of Broadcast Engineering
- Vol. 31 (2) , 227-239
3D Face Reconstruction from a Single RGB Image via Normal-to-3DMM Regression
- 31 March 2026
- Published by The Korean Institute of Broadcast and Media Engineers in Journal of Broadcast Engineering
- Vol. 31 (2) , 227-239
Open Access3D Face Reconstruction from a Single RGB Image via Normal-to-3DMM Regression
- 25 March 2026
- Published by Elsevier in Pattern Recognition
- Vol. 179
Bidirectional regression for monocular 6DoF head pose estimation and reference system alignment
- 25 March 2026
- Published by Elsevier in Pattern Recognition
- Vol. 179
Bidirectional regression for monocular 6DoF head pose estimation and reference system alignment
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 10044-10053
Stable Diffusion-Based Approach for Human De-Occlusion
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 10044-10053
Open AccessStable Diffusion-Based Approach for Human De-Occlusion
- Open Access
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 174219-174235
IVAH: Invisible-Region Augmented Human Rendering From Monocular Images
Most neural radiance field (NeRF)-based human rendering methods rely on per-subject optimization or multi-view data input to reconstruct the 3D radiance field of human subjects. However, optimization-based approaches lack the flexibility to render multiple subjects with a single model, while acquiring multi-view data necessitates a calibrated multi-view setup. To overcome these limitations, monocular image-based methods have been introduced to enable the rendering of novel views and poses of human subjects in a feed-forward manner from a single image. Nonetheless, monocular images inherently provide limited information, presenting challenges when rendering occluded or invisible regions of the subject. To address this limitation, we propose InVisible-region Augmented Human rendering (IVAH), a method that generates and leverages additional appearance data to represent the unseen regions of the input observation image. Specifically, IVAH synthesizes a back image and a UV map of the human subject, extracting features that facilitate the rendering of the occluded areas. Experimental results demonstrate that the inclusion of generated appearance data significantly enhances the rendering of invisible regions. Furthermore, our findings indicate that IVAH achieves superior performance over current monocular image-based techniques.
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 174219-174235
Open AccessIVAH: Invisible-Region Augmented Human Rendering From Monocular Images
Most neural radiance field (NeRF)-based human rendering methods rely on per-subject optimization or multi-view data input to reconstruct the 3D radiance field of human subjects. However, optimization-based approaches lack the flexibility to render multiple subjects with a single model, while acquiring multi-view data necessitates a calibrated multi-view setup. To overcome these limitations, monocular image-based methods have been introduced to enable the rendering of novel views and poses of human subjects in a feed-forward manner from a single image. Nonetheless, monocular images inherently provide limited information, presenting challenges when rendering occluded or invisible regions of the subject. To address this limitation, we propose InVisible-region Augmented Human rendering (IVAH), a method that generates and leverages additional appearance data to represent the unseen regions of the input observation image. Specifically, IVAH synthesizes a back image and a UV map of the human subject, extracting features that facilitate the rendering of the occluded areas. Experimental results demonstrate that the inclusion of generated appearance data significantly enhances the rendering of invisible regions. Furthermore, our findings indicate that IVAH achieves superior performance over current monocular image-based techniques.
- Open Access
Stable Diffusion-Based Approach for Human De-Occlusion
Humans can infer the missing parts of an occluded object by leveraging prior knowledge and visible cues. However, enabling deep learning models to accurately predict such occluded regions remains a challenging task. De-occlusion addresses this problem by reconstructing both the mask and RGB appearance. In this work, we focus on human de-occlusion, specifically targeting the recovery of occluded body structures and appearances. Our approach decomposes the task into two stages: mask completion and RGB completion. The first stage leverages a diffusion-based human body prior to provide a comprehensive representation of body structure, combined with occluded joint heatmaps that offer explicit spatial cues about missing regions. The reconstructed amodal mask then serves as a conditioning input for the second stage, guiding the model on which areas require RGB reconstruction. To further enhance RGB generation, we incorporate human-specific textual features derived using a visual question answering (VQA) model and encoded via a CLIP encoder. RGB completion is performed using Stable Diffusion, with decoder fine-tuning applied to mitigate pixel-level degradation in visible regions -- a known limitation of prior diffusion-based de-occlusion methods caused by latent space transformations. Our method effectively reconstructs human appearances even under severe occlusions and consistently outperforms existing methods in both mask and RGB completion. Moreover, the de-occluded images generated by our approach can improve the performance of downstream human-centric tasks, such as 2D pose estimation and 3D human reconstruction. The code will be made publicly available.
Open AccessStable Diffusion-Based Approach for Human De-Occlusion
Humans can infer the missing parts of an occluded object by leveraging prior knowledge and visible cues. However, enabling deep learning models to accurately predict such occluded regions remains a challenging task. De-occlusion addresses this problem by reconstructing both the mask and RGB appearance. In this work, we focus on human de-occlusion, specifically targeting the recovery of occluded body structures and appearances. Our approach decomposes the task into two stages: mask completion and RGB completion. The first stage leverages a diffusion-based human body prior to provide a comprehensive representation of body structure, combined with occluded joint heatmaps that offer explicit spatial cues about missing regions. The reconstructed amodal mask then serves as a conditioning input for the second stage, guiding the model on which areas require RGB reconstruction. To further enhance RGB generation, we incorporate human-specific textual features derived using a visual question answering (VQA) model and encoded via a CLIP encoder. RGB completion is performed using Stable Diffusion, with decoder fine-tuning applied to mitigate pixel-level degradation in visible regions -- a known limitation of prior diffusion-based de-occlusion methods caused by latent space transformations. Our method effectively reconstructs human appearances even under severe occlusions and consistently outperforms existing methods in both mask and RGB completion. Moreover, the de-occluded images generated by our approach can improve the performance of downstream human-centric tasks, such as 2D pose estimation and 3D human reconstruction. The code will be made publicly available.
- Open Access
- 31 May 2025
- Published by The Korean Institute of Broadcast and Media Engineers in Journal of Broadcast Engineering
- Vol. 30 (3) , 437-448
One-Stage Method for Real-time 3D WholeBody Pose Estimation from Egocentric Images
- 31 May 2025
- Published by The Korean Institute of Broadcast and Media Engineers in Journal of Broadcast Engineering
- Vol. 30 (3) , 437-448
Open AccessOne-Stage Method for Real-time 3D WholeBody Pose Estimation from Egocentric Images
- 25 October 2024
- Published by Springer Nature
- p. 146-163
6DoF Head Pose Estimation Through Explicit Bidirectional Interaction with Face Geometry
- 25 October 2024
- Published by Springer Nature
- p. 146-163
6DoF Head Pose Estimation Through Explicit Bidirectional Interaction with Face Geometry
- Open Access
- 30 November 2023
- Published by The Korean Institute of Broadcast and Media Engineers in Journal of Broadcast Engineering
- Vol. 28 (6) , 698-708
Adaptive Multi-person Tracking Method considering Appearance Changes
- 30 November 2023
- Published by The Korean Institute of Broadcast and Media Engineers in Journal of Broadcast Engineering
- Vol. 28 (6) , 698-708
Open AccessAdaptive Multi-person Tracking Method considering Appearance Changes
- Open Access
Representation learning of vertex heatmaps for 3D human mesh reconstruction from multi-view images
This study addresses the problem of 3D human mesh reconstruction from multi-view images. Recently, approaches that directly estimate the skinned multi-person linear model (SMPL)-based human mesh vertices based on volumetric heatmap representation from input images have shown good performance. We show that representation learning of vertex heatmaps using an autoencoder helps improve the performance of such approaches. Vertex heatmap autoencoder (VHA) learns the manifold of plausible human meshes in the form of latent codes using AMASS, which is a large-scale motion capture dataset. Body code predictor (BCP) utilizes the learned body prior from VHA for human mesh reconstruction from multi-view images through latent code-based supervision and transfer of pretrained weights. According to experiments on Human3.6M and LightStage datasets, the proposed method outperforms previous methods and achieves state-of-the-art human mesh reconstruction performance.
Open AccessRepresentation learning of vertex heatmaps for 3D human mesh reconstruction from multi-view images
This study addresses the problem of 3D human mesh reconstruction from multi-view images. Recently, approaches that directly estimate the skinned multi-person linear model (SMPL)-based human mesh vertices based on volumetric heatmap representation from input images have shown good performance. We show that representation learning of vertex heatmaps using an autoencoder helps improve the performance of such approaches. Vertex heatmap autoencoder (VHA) learns the manifold of plausible human meshes in the form of latent codes using AMASS, which is a large-scale motion capture dataset. Body code predictor (BCP) utilizes the learned body prior from VHA for human mesh reconstruction from multi-view images through latent code-based supervision and transfer of pretrained weights. According to experiments on Human3.6M and LightStage datasets, the proposed method outperforms previous methods and achieves state-of-the-art human mesh reconstruction performance.
- Open Access
Camera Motion Agnostic Method for Estimating 3D Human Poses
Although the performance of 3D human pose and shape estimation methods has improved considerably in recent years, existing approaches typically generate 3D poses defined in a camera or human-centered coordinate system. This makes it difficult to estimate a person’s pure pose and motion in a world coordinate system for a video captured using a moving camera. To address this issue, this paper presents a camera motion agnostic approach for predicting 3D human pose and mesh defined in the world coordinate system. The core idea of the proposed approach is to estimate the difference between two adjacent global poses (i.e., global motion) that is invariant to selecting the coordinate system, instead of the global pose coupled to the camera motion. To this end, we propose a network based on bidirectional gated recurrent units (GRUs) that predicts the global motion sequence from the local pose sequence consisting of relative rotations of joints called global motion regressor (GMR). We use 3DPW and synthetic datasets, which are constructed in a moving-camera environment, for evaluation. We conduct extensive experiments and prove the effectiveness of the proposed method empirically.
Open AccessCamera Motion Agnostic Method for Estimating 3D Human Poses
Although the performance of 3D human pose and shape estimation methods has improved considerably in recent years, existing approaches typically generate 3D poses defined in a camera or human-centered coordinate system. This makes it difficult to estimate a person’s pure pose and motion in a world coordinate system for a video captured using a moving camera. To address this issue, this paper presents a camera motion agnostic approach for predicting 3D human pose and mesh defined in the world coordinate system. The core idea of the proposed approach is to estimate the difference between two adjacent global poses (i.e., global motion) that is invariant to selecting the coordinate system, instead of the global pose coupled to the camera motion. To this end, we propose a network based on bidirectional gated recurrent units (GRUs) that predicts the global motion sequence from the local pose sequence consisting of relative rotations of joints called global motion regressor (GMR). We use 3DPW and synthetic datasets, which are constructed in a moving-camera environment, for evaluation. We conduct extensive experiments and prove the effectiveness of the proposed method empirically.