LPLong Hoang Pham

Long Hoang Pham

*This is an algorithmically generated author record based on publicly available information.

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


Long Hoang Pham has a total of 27 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Dual Memory Networks Guided Reverse Distillation for Unsupervised Anomaly Detection

    by Chi Dai Tran, Long Hoang Pham, Duong Nguyen-Ngoc Tran, Quoc Pham-Nam Ho, Jae Wook Jeon


  • Dynamic Retraining-Updating Mean Teacher for Source-Free Object Detection

    by Trinh Le Ba Khanh, Huy-Hung Nguyen, Long Hoang Pham, Duong Nguyen-Ngoc Tran, Jae Wook Jeon


  • Open Access

    Dynamic Retraining-Updating Mean Teacher for Source-Free Object Detection

    by Trinh Le Ba Khanh, Huy-Hung Nguyen, Long Hoang Pham, Duong Nguyen-Ngoc Tran, Jae Wook Jeon

    In object detection, unsupervised domain adaptation (UDA) aims to transfer knowledge from a labeled source domain to an unlabeled target domain. However, UDA's reliance on labeled source data restricts its adaptability in privacy-related scenarios. This study focuses on source-free object detection (SFOD), which adapts a source-trained detector to an unlabeled target domain without using labeled source data. Recent advancements in self-training, particularly with the Mean Teacher (MT) framework, show promise for SFOD deployment. However, the absence of source supervision significantly compromises the stability of these approaches. We identify two primary issues, (1) uncontrollable degradation of the teacher model due to inopportune updates from the student model, and (2) the student model's tendency to replicate errors from incorrect pseudo labels, leading to it being trapped in a local optimum. Both factors contribute to a detrimental circular dependency, resulting in rapid performance degradation in recent self-training frameworks. To tackle these challenges, we propose the Dynamic Retraining-Updating (DRU) mechanism, which actively manages the student training and teacher updating processes to achieve co-evolutionary training. Additionally, we introduce Historical Student Loss to mitigate the influence of incorrect pseudo labels. Our method achieves state-of-the-art performance in the SFOD setting on multiple domain adaptation benchmarks, comparable to or even surpassing advanced UDA methods. The code will be released at https://github.com/lbktrinh/DRU


  • Open Access

    Optimizing Monocular Driving Assistance for Real-Time Processing on Jetson AGX Xavier

    by Huy-Hung Nguyen, Duong Nguyen-Ngoc Tran, Long Hoang Pham, Jae Wook Jeon

    While computer vision and computing technology advances have facilitated advanced driver assistance applications, systems with multi-task design remain highly demanding to operate at high speed on resource-constrained devices. Our study addresses this challenge by proposing a real-time driver assistance solution specifically developed for a single Jetson AGX Xavier embedded device. It simultaneously performs lane detection, traffic object detection and recognition, and rule-based scene analysis. To achieve high throughput (up to 43 frames per second) without reliance on additional hardware or cloud server, the system exploits Jetson device’s specialized AI accelerator and employs various optimization techniques: multithreaded programming, the TensorRT framework, and post-training quantization. The modular design integrates state-of-the-art task-specific methods and ensures adaptation to diverse traffic scenarios across countries as well as future hardware and solutions. Experimental results using a Korean dashcam traffic dataset validated the system’s effectiveness and practicality.


  • A Vision-Based Method for Real-Time Traffic Flow Estimation on Edge Devices

    by Duong Nguyen-Ngoc Tran, Long Hoang Pham, Huy-Hung Nguyen, Jae Wook Jeon

    Traffic flow estimation is an essential task in modern intelligent transportation systems. Many types of information, including vehicle type, vehicle totals, and movement direction, are vital for mitigating transportation-related tasks and effective traffic control strategies. With the development of embedded devices, systems can process captured video at the edge instead of transferring data to centralized processing servers. This paper proposes a real-time and edge-based traffic flow estimation system. The proposed system follows a detect-and-track mechanism where lightweight deep learning models perform vehicle detection. A novel scenario-based tracking and counting technique is developed to provide multi-class, multi-movement vehicle counting. The method uses predefined regions to assign the movement for each vehicle initially detected. It then performs spatial-temporal trajectory matching between the vehicle trajectory and the movement path throughout the whole video. Extensive experiments have shown that the proposed method achieves high effectiveness with multiple camera types and viewpoints.


  • Open Access

    S2TD: a Separation Logic Verifier that Supports Reasoning of the Absence and Presence of Bugs

    by Quang Loc Le, Jun Sun, Long H. Pham, Shengchao Qin

    Heap-manipulating programs are known to be challenging to reason about. We present a novel verifier for heap-manipulating programs called S2TD, which encodes programs systematically in the form of Constrained Horn Clauses (CHC) using a novel extension of separation logic (SL) with recursive predicates and dangling predicates. S2TD actively explores cyclic proofs to address the path explosion problem. S2TD differentiates itself from existing CHC-based verifiers by focusing on heap-manipulating programs and employing cyclic proof to efficiently verify or falsify them with counterexamples. Compared with existing SL-based verifiers, S2TD precisely specifies the heaps of de-allocated pointers to avoid false positives in reasoning about the presence of bugs. S2TD has been evaluated using a comprehensive set of benchmark programs from the SV-COMP repository. The results show that S2TD is more effective than state-of-art program verifiers and is more efficient than most of them.

  • Open Access

    Universal Detection-Based Driving Assistance Using a Mono Camera With Jetson Devices

    by Duong Nguyen-Ngoc Tran, Long Hoang Pham, Huy-Hung Nguyen, Tai Huu-Phuong Tran, Hyung-Joon Jeon, Jae Wook Jeon

    Advanced Driver Assistance Systems (ADAS) are a collection of intelligent solutions integrated into next-generation vehicles to assist in safe driving. When building ADAS systems, the main goals are that they are stable, flexible, easy to maintain, and allow for error tracing. If a driving assistance algorithm is designed to be implemented on one machine or in one model, there is a potential disadvantage that if one component fails, then the entire system would stop. We work on modularizing the ADAS system to be flexible to accommodate any changes or improvements based on up-to-date requirements. Using advanced current edge (or network) devices, we propose a Detection-based Driving Assistance algorithm, which can collaborate or integrate with an existing system in a vehicle. The core of any process is to ensure that the system has a predictable level of functionality and that any misbehavior can be easily traced to the root cause. The proposed system shows fast, real-time performance on edge devices with limited computing power.


  • Open Access

    Verifying Neural Networks Against Backdoor Attacks

    by Long H. Pham, Jun Sun

    Neural networks have achieved state-of-the-art performance in solving many problems, including many applications in safety/security-critical systems. Researchers also discovered multiple security issues associated with neural networks. One of them is backdoor attacks, i.e., a neural network may be embedded with a backdoor such that a target output is almost always generated in the presence of a trigger. Existing defense approaches mostly focus on detecting whether a neural network is 'backdoored' based on heuristics, e.g., activation patterns. To the best of our knowledge, the only line of work which certifies the absence of backdoor is based on randomized smoothing, which is known to significantly reduce neural network performance. In this work, we propose an approach to verify whether a given neural network is free of backdoor with a certain level of success rate. Our approach integrates statistical sampling as well as abstract interpretation. The experiment results show that our approach effectively verifies the absence of backdoor or generates backdoor triggers.

  • Analysis of the Smartphone Camera Exposure Effect on Laser Extraction

    by Long Hoang Pham, Duong Nguyen-Ngoc Tran, Chul Hong Rhie, Jae Wook Jeon

    In this study, we investigate the effect of smartphone camera exposure on laser extraction performance in a digital image. The laser extraction is the most crucial operation in laser triangulation systems that obtain object measurements from the projected laser light captured in the image. The exposure settings consist of sensor sensitivity (ISO), aperture, and shutter speed. This study provides insight into the smartphone settings and the exposure adjustment for further complicated laser extraction. A smartphone camera (Apple iPhone 11) and a laser stripe projector operating at 405 nm were employed. Analysis of different ISO values, the effect of aperture and shutter speed on the laser extraction quality was performed.