修士課程2年 伊藤君が中国・蘭州で開催されたAIoTSys2025で口頭発表を行い，Best Paper Awardを受賞しました．
Takashi Ito, Wenhao Huang, Yin Chen, Jin Nakazawa. (2025, August). Object Size Classification in Garbage Disposal Sensing System Using Monocular Depth Estimation. In 2025 International Conference on Artificial Intelligence of Things and Systems (AIoTSys). IEEE.
博士課程3年 黄君の研究「AdaLine: Adaptive Counting Line Optimization by Perspective-Aware Trajectory Modeling in Object-Detection-Tracking Systems」が国際論文誌 IEEE Accessにacceptされました．
Abstract:
The combination of object detection, tracking, and counting has become a widely used method. The position and angle of cameras can vary according to the deployment scenarios, which affects counting accuracy. Traditional approaches often rely on manually pre-defined counting lines or regions-of-interest (ROIs), which are static, environment-specific, and difficult to generalize. To overcome these limitations, we propose AdaLine, an adaptive, perspective-aware algorithm that learns the optimal counting line from object trajectories, thereby enabling adaptation to diverse environments and camera viewpoints without manually defined counting lines. AdaLine adapts automatically as the scene evolves by clustering incoming trajectories with K-means, selecting the most stable line candidate, and smoothing it with an exponential moving average. Experimental evaluations across different scenarios and camera settings show that AdaLine achieves better performance in terms of accuracy, stability, and applicability. Our approach offers a scalable, real-time configuration-free solution for object-detection-tracking systems.
修士課程2年 伊藤君の研究「Object Size Classification in Garbage Disposal Sensing System Using Monocular Depth Estimation」が国際会議 AIoTSys 2025にacceptされました．
Abstract:
Deep learning-based object detection is widely used in urban sensing, enabling tasks such as pedestrian, pothole, and waste detection. Automotive sensing with dashcams facili- tates large-scale, real-time detection across urban environments. However, existing studies primarily focus on detection without estimating object size, which is crucial for event classification. Conventional size estimation methods rely on RGB-D cameras, multiple cameras, or LIDAR, making them unsuitable for large- scale automotive sensing with single RGB dashcams. Monocular depth estimation provides relative depth but does not yield abso- lute size measurements. To address this limitation, we propose a novel approach that combines monocular depth estimation with a reference object of known size. By comparing the detected object’s pixel dimensions with those of the reference object, its physical size can be estimated. To validate our approach, we developed an automotive sensing platform that detects and quantifies household garbage bags using footage from the rear- view camera of garbage trucks. The truck body serves as the reference object, ensuring reliable size estimation. Experiments conducted with real-world data collected using an NVIDIA Jetson TX2 demonstrate the effectiveness of our method. The proposed approach achieves size estimation accuracy with mean squared errors (MSEs) of 20.02 for width and 18.68 for height while maintaining an end-to-end processing rate of 19.21 frames per second (FPS) for detection, tracking, and size estimation.
後期博士課程 三上君の研究「JumpQ: Stochastic Scheduling to Accelerating Object-detection-driven Mobile Sensing on Object-sparse Video Data」が国際会議 ACM SenSys2025にacceptされ，口頭発表を行いました．
Abstract:
Deep learning-based object detection has seen a surge in application for sensing systems on mobile devices. In this context, objects are identified and tracked across video frames, facilitating the calculation of associated events of interest. A significant research challenge refers to the acceleration of processing speed, which is constrained by deep learningbased object detection due to its intensive resource requirements. This paper focuses on a typical mobile sensing scenario, wherein sequences of frames containing objects of interest are sparsely dispersed throughout the video stream. Given that many of the frames lack objects, allocating substantial computational resources to detect them becomes inefficient. In light of this, we propose a stochastic scheduling algorithm, JumpQ. JumpQ performs per-frame detection when anticipating the presence of objects in the current frames. Consecutive negative detections prompt a transition to intermittent detection with a probability that undergoes further decay if the negative detection persists until reaching a predefined limit. Upon a positive detection, JumpQ swiftly reverts to per-frame detection and retraces a specific number of previously buffered frames to ensure the inclusion of potentially missed true frames. A comprehensive experimental study using the garbage bag counting technique was conducted to show the efficiency of JumpQ in accelerating the processing speed by nearly 1.92 times while maintaining a negligible impact on sensing accuracy.
Mikami, K., Huang, W., Chen, Y., & Nakazawa, J. (2025, May). JumpQ: Stochastic Scheduling to Accelerating Object-detection-driven Mobile Sensing on Object-sparse Video Data. In Proceedings of the 23rd ACM Conference on Embedded Networked Sensor Systems (pp. 332-344).
卒業生 牧野君の研究「Laqista: Serverless Cloud-Fog-Dew Computing Platform for Deep Learning Applications」が国際会議 IEEE SmartComp2025にacceptされました．
Abstract:
In Smart Things and smart city applications, IoT devices generate large amounts of data and deep learning technologies are used to acquire useful information from it. Based on the kind of application and data, there are various non-functional requirements, such as low latency for information presentation by MR and privacy for video processing. To serve these requirements, a computing platform needs to make appropriate use of computing resources, namely Cloud, Fog, and Dew. However, there are some technical challenges in designing such a platform: i) transparently satisfying application QoS; ii) running the application across various hardware and OSes without modification; iii) sharing the application context taking into account the validity of values (temporal locality) and the data privacy (spatial locality). In this paper, we introduce Laqista, a novel Cloud-Fog-Dew computing platform. Laqista serves applications in a serverless manner via the Edgeless API, which schedules requests and abstracts the details of the platform. Applications are separated into Logics and Models, which are converted to lightweight, platform-agnostic formats such as WebAssembly and ONNX, respectively. Additionally, the Context Store synchronizes application context among the nodes, handling the privacy and validity of data. We developed a prototype implementation of Laqista in Rust and evaluated its performance. Experimental results show that the Laqista design has practical performance and is applicable to real-time applications such as video processing and MR.
卒業生 高君らの論文「Cyberoception: Finding A Painlessly-Measurable New Sense In The Cyberworld Towards Emotion-awareness In Computing」がACM CHI’25に採択されました。学部4年次の研究です。
Abstract:
In Affective computing, recognizing users’ emotions accurately is the basis of affective human-computer interaction. Understanding users’ interoception contributes to a better understanding of individually different emotional abilities, which is essential for achieving inter-individually accurate emotion estimation. However, existing interoception measurement methods, such as the heart rate discrimination task, have several limitations, including their dependence on a well-controlled laboratory environment and precision apparatus, making monitoring users’ interoception challenging. This study aims to determine other forms of data that can explain users’ interoceptive or similar states in their real-world lives and propose a novel hypothetical concept “cyberoception,” a new sense (1) which has properties similar to interoception in terms of the correlation with other emotion-related abilities, and (2) which can be measured only by the sensors embedded inside commodity smartphone devices in users’ daily lives. Results from a 10-day-long in-lab/in-the-wild hybrid experiment reveal a specific cyberoception type “Turn On” (users’ subjective sensory perception about the frequency of turning-on behavior on their smartphones), significantly related to participants’ emotional valence. We anticipate that cyberoception to serve as a fundamental building block for developing more “emotion-aware”, user-friendly applications and services.
Okoshi, T., Gao, Z., Tan, Y. Z., Karasawa, T., Miki, T., Sasaki, W., & Balan, R. K. (2025, April). Cyberoception: Finding A Painlessly-Measurable New Sense In The Cyberworld Towards Emotion-awareness In Computing. In Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems (pp. 1-17).
