Hojung Cha
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Hojung Cha has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
- Jiwon Kim
Uppsala University
13 shared publications - Jeho Lee
Yonsei University, Seoul, Republic of Korea
10 shared publications - Seunghyeok Jeon
Yonsei University, Seoul, Republic of Korea
8 shared publications - Junick Ahn
Department of Computer Science, Yonsei University, Seoul, South Korea
7 shared publications Rhan Ha
Department of Computer Engineering, Hongik University, Seoul, South Korea
7 shared publications- Seonghoon Park
Yonsei University, Seoul, Republic of Korea
7 shared publications
Latest Publications
- Open Access
- 26 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1812-1828
viNPU: Optimizing Vision Transformer Inference on Mobile NPUs
- 26 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1812-1828
Open AccessviNPU: Optimizing Vision Transformer Inference on Mobile NPUs
- Open Access
- 6 March 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 25 (2) , 1-23
Phoenix: Thermal-Aware On-Device Inference of Multi-Instance DNNs for Mobile Video Applications
Running multiple deep neural networks (DNNs) simultaneously on mobile devices introduces challenges due to constrained computing resources. Previous research has explored the use of heterogeneous processors for accelerating DNN inference but often overlooks thermal issues, which can degrade computing power. In this article, we propose Phoenix, a system specifically designed to enhance the performance of multi-instance DNNs in video applications by maximizing accuracy and ensuring the achievement of a required frame rate. Phoenix allocates DNN tasks to the most suitable hardware processors, understanding complex thermal dynamics through reinforcement learning, and postpones the onset of thermal throttling. Despite optimized task allocation, continuous inference of multiple DNNs can still lead to thermal throttling. To manage performance degradation, Phoenix employs a multi-exit network, adaptively executing inference tasks to ensure consistent frame rates. Phoenix minimizes accuracy loss from early exits by optimally generating and operating multi-exit networks. We evaluated Phoenix using two different benchmarks and Virtual Youtuber streaming application. The results demonstrated that Phoenix effectively enhances device performance by delaying thermal throttling and achieving optimal accuracy while maintaining a consistent frame rate.
- 6 March 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 25 (2) , 1-23
Open AccessPhoenix: Thermal-Aware On-Device Inference of Multi-Instance DNNs for Mobile Video Applications
Running multiple deep neural networks (DNNs) simultaneously on mobile devices introduces challenges due to constrained computing resources. Previous research has explored the use of heterogeneous processors for accelerating DNN inference but often overlooks thermal issues, which can degrade computing power. In this article, we propose Phoenix, a system specifically designed to enhance the performance of multi-instance DNNs in video applications by maximizing accuracy and ensuring the achievement of a required frame rate. Phoenix allocates DNN tasks to the most suitable hardware processors, understanding complex thermal dynamics through reinforcement learning, and postpones the onset of thermal throttling. Despite optimized task allocation, continuous inference of multiple DNNs can still lead to thermal throttling. To manage performance degradation, Phoenix employs a multi-exit network, adaptively executing inference tasks to ensure consistent frame rates. Phoenix minimizes accuracy loss from early exits by optimally generating and operating multi-exit networks. We evaluated Phoenix using two different benchmarks and Virtual Youtuber streaming application. The results demonstrated that Phoenix effectively enhances device performance by delaying thermal throttling and achieving optimal accuracy while maintaining a consistent frame rate.
- Open Access
- 3 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 878-893
EOS: Energy-Optimized Super-Resolution on Mobile Devices for Live 360-Degree Videos
- 3 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 878-893
Open AccessEOS: Energy-Optimized Super-Resolution on Mobile Devices for Live 360-Degree Videos
- Open Access
- 3 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1106-1120
Vega: Fully Immersive Mobile Volumetric Video Streaming with 3D Gaussian Splatting
- 3 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1106-1120
Open AccessVega: Fully Immersive Mobile Volumetric Video Streaming with 3D Gaussian Splatting
- Open Access
- 9 October 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (6) , 1-26
Duration-Aware Sound Event Detection on Ultra-Low-Power Sensor Devices
Sound event detection (SED) based on on-device machine learning (ML) presents considerable energy challenges for ultra-low-power sensor devices. In this article, we propose DASH, a duration-aware SED system designed for energy-constrained sensor devices in domestic environments. As repeated inferences for continuous sound events lead to unnecessary energy consumption, DASH aims to minimize unnecessary inferences by predicting the duration of sound events. However, the variability of sound event durations across different environments and scenarios poses a major challenge in developing a responsive yet energy-efficient duration-aware SED system. To address this, DASH introduces three key solutions: (1) N-probability distribution-based event duration prediction, which identifies checkpoints where new inferences are likely needed; (2) Affinity-guided event classification, which performs low-energy affinity matching at checkpoints to determine whether ML inference is necessary; and (3) Interrupt blocking-enabling cycle-based device state control, which periodically checks for event presence with minimal energy consumption at non-checkpoint times. We implemented DASH on MSP430-based sensor devices deployed in real home environments. Experimental results demonstrate that DASH reduced energy consumption by approximately 97–98% compared to evaluation baselines, with only a 4.7% error rate.
- 9 October 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (6) , 1-26
Open AccessDuration-Aware Sound Event Detection on Ultra-Low-Power Sensor Devices
Sound event detection (SED) based on on-device machine learning (ML) presents considerable energy challenges for ultra-low-power sensor devices. In this article, we propose DASH, a duration-aware SED system designed for energy-constrained sensor devices in domestic environments. As repeated inferences for continuous sound events lead to unnecessary energy consumption, DASH aims to minimize unnecessary inferences by predicting the duration of sound events. However, the variability of sound event durations across different environments and scenarios poses a major challenge in developing a responsive yet energy-efficient duration-aware SED system. To address this, DASH introduces three key solutions: (1) N-probability distribution-based event duration prediction, which identifies checkpoints where new inferences are likely needed; (2) Affinity-guided event classification, which performs low-energy affinity matching at checkpoints to determine whether ML inference is necessary; and (3) Interrupt blocking-enabling cycle-based device state control, which periodically checks for event presence with minimal energy consumption at non-checkpoint times. We implemented DASH on MSP430-based sensor devices deployed in real home environments. Experimental results demonstrate that DASH reduced energy consumption by approximately 97–98% compared to evaluation baselines, with only a 4.7% error rate.
- Open Access
- 26 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (5s) , 1-24
SecureRide: Detecting Safety-Threatening Behavior of E-Scooters Using Battery Information
Reckless usage of electric (e-) scooters causes many injury accidents, raising critical safety concerns. Despite newly introduced regulations, specifically, speed limits and sidewalk driving prohibitions, the number of accidents increases due to the challenges in enforcement. Therefore, a reliable method to detect safety-threatening illegal behaviors of e-scooters is essential to mitigate this growing problem. In this paper, we propose SecureRide, a system that accurately detects illegal e-scooter behaviors, i.e., speeding violation and sidewalk riding, at runtime using only battery information, without the need for additional sensors. To this end, we first design a neural network-based illegal behavior predictor that takes sequences of three battery factors, i.e., voltage, current, and capacity, as inputs. The model architecture is optimized based on time constraints, target accuracy, and resource constraints of the target devices. Next, we devise a runtime detection strategy to achieve both high accuracy and low detection time. SecureRide operates in two modes with different predictors— lightweight-quick and complex-accurate models—depending on the driving situation, ensuring both high accuracy and low detection time. We extensively validate SecureRide based on actual driving experiments. Our results show that SecureRide detects illegal behaviors with an accuracy of up to 99.77% within 1.01 seconds.
- 26 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (5s) , 1-24
Open AccessSecureRide: Detecting Safety-Threatening Behavior of E-Scooters Using Battery Information
Reckless usage of electric (e-) scooters causes many injury accidents, raising critical safety concerns. Despite newly introduced regulations, specifically, speed limits and sidewalk driving prohibitions, the number of accidents increases due to the challenges in enforcement. Therefore, a reliable method to detect safety-threatening illegal behaviors of e-scooters is essential to mitigate this growing problem. In this paper, we propose SecureRide, a system that accurately detects illegal e-scooter behaviors, i.e., speeding violation and sidewalk riding, at runtime using only battery information, without the need for additional sensors. To this end, we first design a neural network-based illegal behavior predictor that takes sequences of three battery factors, i.e., voltage, current, and capacity, as inputs. The model architecture is optimized based on time constraints, target accuracy, and resource constraints of the target devices. Next, we devise a runtime detection strategy to achieve both high accuracy and low detection time. SecureRide operates in two modes with different predictors— lightweight-quick and complex-accurate models—depending on the driving situation, ensuring both high accuracy and low detection time. We extensively validate SecureRide based on actual driving experiments. Our results show that SecureRide detects illegal behaviors with an accuracy of up to 99.77% within 1.01 seconds.
- Open Access
- 26 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (5s) , 1-24
Ember: Task Wakeup Sequence–Based Energy Optimization for Mobile Web Browsing
Existing Android systems exhibit energy inefficiency during mobile web browsing due to the lack of awareness of application-level context. Inferring such context from system-level data alone is challenging, but one promising opportunity is using the sequence of task wakeup events, where one task activates another. These sequences show correlation with the type of webpage being used. In this paper, we present Ember, a lightweight and responsive power management system for mobile web browsing using only task wakeup sequences. Ember introduces a neural network–based approach to predict optimal CPU clamping values by addressing three key challenges: (1) embedding task names, given as natural-language strings, into meaningful vectors using a Word2Vec-based embedding scheme tailored for task wakeup sequences; (2) minimizing inference overhead with a touch-driven hierarchical inference method that combines lightweight logistic regression with high-accuracy neural networks to balance responsiveness and efficiency; and (3) adapting to within-page interaction dynamics through an interaction-adaptive clamping mechanism that adjusts constraints across different user interaction phases. Implemented on commercial Android smartphones, Ember reduced power consumption by 6.2%–31.2% across a wide range of webpages while maintaining user-perceived quality of experience (QoE).
- 26 September 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 24 (5s) , 1-24
Open AccessEmber: Task Wakeup Sequence–Based Energy Optimization for Mobile Web Browsing
Existing Android systems exhibit energy inefficiency during mobile web browsing due to the lack of awareness of application-level context. Inferring such context from system-level data alone is challenging, but one promising opportunity is using the sequence of task wakeup events, where one task activates another. These sequences show correlation with the type of webpage being used. In this paper, we present Ember, a lightweight and responsive power management system for mobile web browsing using only task wakeup sequences. Ember introduces a neural network–based approach to predict optimal CPU clamping values by addressing three key challenges: (1) embedding task names, given as natural-language strings, into meaningful vectors using a Word2Vec-based embedding scheme tailored for task wakeup sequences; (2) minimizing inference overhead with a touch-driven hierarchical inference method that combines lightweight logistic regression with high-accuracy neural networks to balance responsiveness and efficiency; and (3) adapting to within-page interaction dynamics through an interaction-adaptive clamping mechanism that adjusts constraints across different user interaction phases. Implemented on commercial Android smartphones, Ember reduced power consumption by 6.2%–31.2% across a wide range of webpages while maintaining user-perceived quality of experience (QoE).
- Open Access
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 617-618
Poster: Mixture of Class-aware Experts for Efficient AIoT Inference
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 617-618
Open AccessPoster: Mixture of Class-aware Experts for Efficient AIoT Inference
- Open Access
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 249-262
ARIA: Optimizing Vision Foundation Model Inference on Heterogeneous Mobile Processors for Augmented Reality
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 249-262
Open AccessARIA: Optimizing Vision Foundation Model Inference on Heterogeneous Mobile Processors for Augmented Reality
- Open Access
- 4 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1207-1221
Panopticus: Omnidirectional 3D Object Detection on Resource-constrained Edge Devices
3D object detection with omnidirectional views enables safety-critical applications such as mobile robot navigation. Such applications increasingly operate on resource-constrained edge devices, facilitating reliable processing without privacy concerns or network delays. To enable cost-effective deployment, cameras have been widely adopted as a low-cost alternative to LiDAR sensors. However, the compute-intensive workload to achieve high performance of camera-based solutions remains challenging due to the computational limitations of edge devices. In this paper, we present Panopticus, a carefully designed system for omnidirectional and camera-based 3D detection on edge devices. Panopticus employs an adaptive multi-branch detection scheme that accounts for spatial complexities. To optimize the accuracy within latency limits, Panopticus dynamically adjusts the model's architecture and operations based on available edge resources and spatial characteristics. We implemented Panopticus on three edge devices and conducted experiments across real-world environments based on the public self-driving dataset and our mobile 360° camera dataset. Experiment results showed that Panopticus improves accuracy by 62% on average given the strict latency objective of 33ms. Also, Panopticus achieves a 2.1× latency reduction on average compared to baselines.
- 4 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1207-1221
Open AccessPanopticus: Omnidirectional 3D Object Detection on Resource-constrained Edge Devices
3D object detection with omnidirectional views enables safety-critical applications such as mobile robot navigation. Such applications increasingly operate on resource-constrained edge devices, facilitating reliable processing without privacy concerns or network delays. To enable cost-effective deployment, cameras have been widely adopted as a low-cost alternative to LiDAR sensors. However, the compute-intensive workload to achieve high performance of camera-based solutions remains challenging due to the computational limitations of edge devices. In this paper, we present Panopticus, a carefully designed system for omnidirectional and camera-based 3D detection on edge devices. Panopticus employs an adaptive multi-branch detection scheme that accounts for spatial complexities. To optimize the accuracy within latency limits, Panopticus dynamically adjusts the model's architecture and operations based on available edge resources and spatial characteristics. We implemented Panopticus on three edge devices and conducted experiments across real-world environments based on the public self-driving dataset and our mobile 360° camera dataset. Experiment results showed that Panopticus improves accuracy by 62% on average given the strict latency objective of 33ms. Also, Panopticus achieves a 2.1× latency reduction on average compared to baselines.