Author name cluster

Dongjin Kim

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.

5 papers

2 author rows

AAAI Conference 2026 Conference Paper

Continuous Degradation Modeling via Latent Flow Matching for Real-World Super-Resolution

Hyeonjae Kim
Dongjin Kim
Eugene Jin
Tae Hyun Kim

While deep learning-based super-resolution (SR) methods have shown impressive outcomes with synthetic degradation scenarios such as bicubic downsampling, they frequently struggle to perform well on real-world images that feature complex, nonlinear degradations like noise, blur, and compression artifacts. Recent efforts to address this issue have involved the painstaking compilation of real low-resolution (LR) and high-resolution (HR) image pairs, usually limited to several specific downscaling factors. To address these challenges, our work introduces a novel framework capable of synthesizing authentic LR images from a single HR image by leveraging the latent degradation space with flow matching. Our approach generates LR images with realistic artifacts at unseen degradation levels, which facilitates the creation of large-scale, real-world SR training datasets. Comprehensive quantitative and qualitative assessments verify that our synthetic LR images accurately replicate real-world degradations. Furthermore, both traditional and arbitrary-scale SR models trained using our datasets consistently yield much better HR outcomes.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Watch Video, Catch Keyword: Context-aware Keyword Attention for Moment Retrieval and Highlight Detection

Sung Jin Um
Dongjin Kim
Sangmin Lee
Jung Uk Kim

The goal of video moment retrieval and highlight detection is to identify specific segments and highlights based on a given text query. With the rapid growth of video content and the overlap between these tasks, recent works have addressed both simultaneously. However, they still struggle to fully capture the overall video context, making it challenging to determine which words are most relevant. In this paper, we present a novel Video Context-aware Keyword Attention module that overcomes this limitation by capturing keyword variation within the context of the entire video. To achieve this, we introduce a video context clustering module that provides concise representations of the overall video context, thereby enhancing the understanding of keyword dynamics. Furthermore, we propose a keyword weight detection module with keyword-aware contrastive learning that incorporates keyword information to enhance fine-grained alignment between visual and textual features. Extensive experiments on the QVHighlights, TVSum, and Charades-STA benchmarks demonstrate that our proposed method significantly improves performance in moment retrieval and highlight detection tasks compared to existing approaches.

PDF Details DOI

AAAI Conference 2024 Conference Paper

REPrune: Channel Pruning via Kernel Representative Selection

Mincheol Park
Dongjin Kim
Cheonjun Park
Yuna Park
Gyeong Eun Gong
Won Woo Ro
Suhyun Kim

Channel pruning is widely accepted to accelerate modern convolutional neural networks (CNNs). The resulting pruned model benefits from its immediate deployment on general-purpose software and hardware resources. However, its large pruning granularity, specifically at the unit of a convolution filter, often leads to undesirable accuracy drops due to the inflexibility of deciding how and where to introduce sparsity to the CNNs. In this paper, we propose REPrune, a novel channel pruning technique that emulates kernel pruning, fully exploiting the finer but structured granularity. REPrune identifies similar kernels within each channel using agglomerative clustering. Then, it selects filters that maximize the incorporation of kernel representatives while optimizing the maximum cluster coverage problem. By integrating with a simultaneous training-pruning paradigm, REPrune promotes efficient, progressive pruning throughout training CNNs, avoiding the conventional train-prune-finetune sequence. Experimental results highlight that REPrune performs better in computer vision tasks than existing methods, effectively achieving a balance between acceleration ratio and performance retention.

PDF Details DOI

ICRA Conference 2023 Conference Paper

Control of Shape Memory Alloy Actuator via Electrostatic Capacitive Sensor for Meso-scale Mirror Tilting System

Baekgyeom Kim
Doohoe Lee
Dongjin Kim
Seungyong Han
Daeshik Kang
Uikyum Kim
Je-Sung Koh

Shape memory alloy (SMA) has superior actuation capability over the limit of the scale. However, inherently low controllability is a primary issue that hinders practical usage. To address this challenge, this paper presents an SMA-based artificial muscle actuator capable of the displacement sensing through the capacitive sensor. To realize sensing capability, the theoretical model-based design and fabrication process are proposed. Here, we show that the actuator can be controlled at intervals of 100 μm as well as maintaining sensing capability while lifting 90 times heavier than its weight. To exhibit the usefulness of the actuator to an optical device, we integrate the actuator into the mirror tilting device, which has 20 degrees tilting angle. We expect that the proposed actuator can overcome the scale limit of meso-scale devices, which require payload capacity and controllability, simultaneously.

Details

NeurIPS Conference 2023 Conference Paper

Tanh Works Better with Asymmetry

Dongjin Kim
Woojeong Kim
Suhyun Kim

Batch Normalization is commonly located in front of activation functions, as proposed by the original paper. Swapping the order, i. e. , using Batch Normalization after activation functions, has also been attempted, but its performance is generally not much different from the conventional order when ReLU or a similar activation function is used. However, in the case of bounded activation functions like Tanh, we discovered that the swapped order achieves considerably better performance than the conventional order on various benchmarks and architectures. This paper reports this remarkable phenomenon and closely examines what contributes to this performance improvement. By looking at the output distributions of individual activation functions, not the whole layers, we found that many of them are asymmetrically saturated. The experiments designed to induce a different degree of asymmetric saturation support the hypothesis that asymmetric saturation helps improve performance. In addition, Batch Normalization after bounded activation functions relocates the asymmetrically saturated output of activation functions near zero, enabling the swapped model to have high sparsity, further improving performance. Extensive experiments with Tanh, LeCun Tanh, and Softsign show that the swapped models achieve improved performance with a high degree of asymmetric saturation. Finally, based on this investigation, we test a Tanh function shifted to be asymmetric. This shifted Tanh function that is manipulated to have consistent asymmetry shows even higher accuracy than the original Tanh used in the swapped order, confirming the asymmetry's importance. The code is available at https: //github. com/hipros/tanh works better with asymmetry.

PDF Details