Author name cluster

Minho Park

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.

4 papers

1 author row

AAAI Conference 2026 Conference Paper

SphereDiff: Tuning-free 360° Static and Dynamic Panorama Generation via Spherical Latent Representation

Minho Park
Taewoong Kang
Jooyeol Yun
Sungwon Hwang
Jaegul Choo

The increasing demand for AR/VR applications has highlighted the need for high-quality content, such as 360° live wallpapers. However, generating high-quality 360° panoramic contents remains a challenging task due to the severe distortions introduced by equirectangular projection (ERP). Existing approaches either fine-tune pretrained diffusion models on limited ERP datasets or adopt tuning-free methods that still rely on ERP latent representations, often resulting in distracting distortions near the poles. In this paper, we introduce SphereDiff, a novel approach for synthesizing 360° static and live wallpaper with state-of-the-art diffusion models without additional tuning. We define a spherical latent representation that ensures consistent quality across all perspectives, including near the poles. Then, we extend MultiDiffusion to spherical latent representation and propose a dynamic spherical latent sampling method to enable direct use of pretrained diffusion models. Moreover, we introduce distortion-aware weighted averaging to further improve the generation quality. Our method outperforms existing approaches in generating 360° static and live wallpaper, making it a robust solution for immersive AR/VR applications.

PDF Details DOI

EAAI Journal 2025 Journal Article

Visual object tracking using learnable target-aware token emphasis

Minho Park
Jinjoo Song
Sang Min Yoon

Visual object tracking, which involves tracking the spatial location of a target object either within a single viewpoint or across various camera perspectives, is an important task in computer vision. Deep neural networks, especially vision transformers, typically outperform traditional methods and have thus become the preferred choice for visual object tasks. However, existing visual object tracking frameworks still struggle to adapt to targets with continuously changing appearances within the current frame, as they rely heavily on the static initial target template rather than continuously emphasizing the evolving target features. In this paper, we introduce a visual object tracking network with a learnable target-aware token emphasis, which is composed of vision transformer backbone embedded in the token emphasizer, localization head and target template update decision module. The learnable target-aware token emphasizer and target template update decision modules in the proposed model contribute to stabilizing visual object tracking across various scenarios. This is achieved not only by emphasizing features that have a relationship between the target template and the search region but also by reducing irrelevant features and consistently updating the high-quality target template online during the process. Qualitative and quantitative analyses, including ablation analysis across a diverse set of tracking benchmark datasets, validate the robustness of the proposed tracking framework. The code and trained models are available at https: //github. com/qkdkralsgh/TETrack.

Details DOI

EAAI Journal 2024 Journal Article

Unified spatio-temporal attention mixformer for visual object tracking

Minho Park
Gang-Joon Yoon
Jinjoo Song
Sang Min Yoon

In this paper, we present a unified spatio-temporal attention MixFormer framework for visual object tracking. Within the vision transformer framework, we design a cohesive network consisting of target template and search region feature extraction, cross-attention utilizing spatial and temporal information, and task-specific heads, all operating in an end-to-end manner. Incorporating spatial and temporal attention modules within the network enables simultaneous feature extraction and emphasis, allowing the model to concentrate on target-specific discriminative features despite changes in illumination, occlusion, scale, camera pose, and background clutter. Stacking multiple non-hierarchical blocks allows meaningful features to be extracted while irrelevant features are discarded from the provided target template and search region. The simultaneous spatio-temporal attention module is employed to accentuate target appearance features and alleviate variation in the object state across frame sequences. Qualitative and quantitative analysis, including ablation tests based on various tracking benchmarks, validates the robustness of the proposed tracking methodology.

Details DOI

AAAI Conference 2021 Conference Paper

Visual Comfort Aware-Reinforcement Learning for Depth Adjustment of Stereoscopic 3D Images

Hak Gu Kim
Minho Park
Sangmin Lee
Seongyeop Kim
Yong Man Ro

Depth adjustment aims to enhance the visual experience of stereoscopic 3D (S3D) images, which accompanied with improving visual comfort and depth perception. For a human expert, the depth adjustment procedure is a sequence of iterative decision making. The human expert iteratively adjusts the depth until he is satisfied with the both levels of visual comfort and the perceived depth. In this work, we present a novel deep reinforcement learning (DRL)-based approach for depth adjustment named VCA-RL (Visual Comfort Aware Reinforcement Learning) to explicitly model human sequential decision making in depth editing operations. We formulate the depth adjustment process as a Markov decision process where actions are defined as camera movement operations to control the distance between the left and right cameras. Our agent is trained based on the guidance of an objective visual comfort assessment metric to learn the optimal sequence of camera movement actions in terms of perceptual aspects in stereoscopic viewing. With extensive experiments and user studies, we show the effectiveness of our VCA-RL model on three different S3D databases.

PDF Details