A5023625406- Data Management and Algorithms
- Human Mobility and Location-Based Analysis
- Advanced Memory and Neural Computing
- Traffic Prediction and Management Techniques
- Geographic Information Systems Studies
- Industrial Vision Systems and Defect Detection
- Advanced Image Processing Techniques
- Ferroelectric and Negative Capacitance Devices
- Advanced Vision and Imaging
- Semiconductor materials and devices
- Video Surveillance and Tracking Methods
- Advanced Database Systems and Queries
- VLSI and FPGA Design Techniques
- Transportation Planning and Optimization
- Image Enhancement Techniques
- Fault Detection and Control Systems
- Recommender Systems and Techniques
- Data Mining Algorithms and Applications
- Neuroscience and Neural Engineering
- Embedded Systems Design Techniques
- Transportation and Mobility Innovations
- Advanced Image and Video Retrieval Techniques
- Analog and Mixed-Signal Circuit Design
- Low-power high-performance VLSI design
- Microgrid Control and Optimization
National Chung Hsing University
2024-2025
Feng Chia University
2014-2024
Adrian College
2023
Directorate of Medicinal and Aromatic Plants Research
2023
National Yang Ming Chiao Tung University
2009-2023
University of Würzburg
2023
National Taiwan University
2023
National Yunlin University of Science and Technology
2017-2022
Tennessee State University
2020-2022
University of Notre Dame
2022
This paper reviews the NTIRE 2023 challenge on efficient single-image super-resolution with a focus proposed solutions and results. The aim of this is to devise network that reduces one or several aspects such as runtime, parameters, FLOPs, activations, memory footprint, depth RFDN while at least maintaining PSNR 29.00dB DIV2K validation datasets. had 272 registered participants, 35 teams made valid submissions. They gauge state-of-the-art for super-resolution.
This paper reviews the NTIRE 2023 challenge on image super-resolution (×4), focusing proposed solutions and results. The task of (SR) is to generate a high-resolution (HR) output from corresponding low-resolution (LR) input by leveraging prior information paired LR-HR images. aim obtain network design/solution capable produce high-quality results with best performance (e.g., PSNR). We want explore how high we can achieve regardless computational cost model size FLOPs) data. track was measure...
This work reviews the results of NTIRE 2023 Challenge on Image Shadow Removal. The described set solutions were proposed for a novel dataset, which captures wide range object-light interactions. It consists 1200 roughly pixel aligned pairs real shadow free and affected images, captured in controlled environment. data was white-box setup, using professional equipment lights acquisition sensors. challenge had number 144 participants registered, out 19 teams compared final ranking. extend...
This report introduces two high-quality datasets Flickr360 and ODV360 for omnidirectional image video super-resolution, respectively, reports the NTIRE 2023 challenge on 360° super-resolution. Unlike ordinary 2D images/videos with a narrow field of view, can represent whole scene from all directions in one shot. There exists large gap between image/video both degradation restoration processes. The is held to facilitate development super-resolution by considering their special...
This paper introduces a novel benchmark for efficient up-scaling as part of the NTIRE 2023 Real-Time Image Super-Resolution (RTSR) Challenge, which aimed to upscale images from 720p and 1080p resolution native 4K (×2 ×3 factors) in real-time on commercial GPUs. For this, we use new test set containing diverse ranging digital art gaming photography. We assessed methods devised SR by measuring their runtime, parameters, FLOPs, while ensuring minimum PSNR fidelity over Bicubic interpolation....
This paper presents a review of the NTIRE 2023 challenge on night photography rendering.The goal was to find solutions that process raw camera images taken in nighttime conditions conditions, and thereby produce photo-quality output standard RGB (sRGB) space.Unlike previous year's competition, participants were not provided with large training dataset for target sensor.Instead, this time they given color checker illuminated by known light source.To evaluate results, sufficient number viewers...
We demonstrates a novel Field-Programmable Gate Array (FPGA) structure based on Resistive Random Access Memory (RRAM) system. RRAM is non-volatile memory device which compatible to CMOS Back End of Line (BEOL) process with only 4F <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> area per cell. use 1R system for logic element, Look-Up-Table (LUT), three dimension stacking structure. The proposed 2R routing elements, Switch Block (SB) and...
In the production process from green beans to coffee bean packages, defective removal (or in short, defect removal) is one of most labor-consuming stages, and many companies investigate automation this stage for minimizing human efforts. paper, we propose a deep-learning-based inspection scheme (DL-DBIS), together with GAN (generative-adversarial network)-structured automated labeled data augmentation method (GALDAM) enhancing proposed scheme, so that degree robotic arms can be further...
Virtual try-on tasks aim at synthesizing realistic results by trying target clothes on humans. Most previous works relied the Thin Plate Spline or appearance flows to warp fit human body shapes. However, both approaches cannot handle complex warping, leading over distortion misalignment. Furthermore, there is a critical unaddressed challenge of adjusting clothing sizes for try-on. To tackle these issues, we propose Clothing-Oriented Transformation Try-On Network (COTTON). COTTON leverages...
Computer numerical control (CNC) tool-wear prediction (TWPred) is an important issue in the industry. Recently, researchers have demonstrated that deep-learning models (DLMs) are effective TWPred. However, DLMs ill-suited to small- and medium-scale manufacturers due high computational costs. Methods exist reduce costs of DLMs, but most them depend on overly-complex pruning processes not appropriate for low-end computers used by above manufacturers. Therefore, we developed a lightweight DLM...
For its simple structure, high density, and good scalability, the resistive random access memory (RRAM) has emerged as one of promising candidates for large data storage in computing systems. Moreover, building up RRAM a 3-D stacking structure further boosts advantage array density. Conventionally, multiple bipolar layers are piled vertically separated with isolation material to prevent signal interference between adjacent layers. The process increases fabrication cost brings potential...
Airborne particulate matter 2.5 (PM2.5) can have a profound effect on the health of population. Many researchers been reporting highly accurate numerical predictions based raw PM2.5 data imported directly into deep learning models; however, there is still considerable room for improvement in terms implementation costs due to heavy computational overhead. From perspective environmental science, values given location be attributed local sources as well external sources. Local tend dramatic...
This paper presents an inertial-sensor-based portable instrument and its associated swing motion detection algorithm to detect the following several major stages of a full golf motion: address, back swing, top down impact, follow through, finish. The consisting microcontroller, triaxial accelerometer, gyroscope, Bluetooth wireless transmission module can be mounted on club measure signals. A composed procedures signal collection, preprocessing, segmentation has been developed for recognizing...
This paper presents low-voltage low-power limiter, RSSI, and demodulator designs for a low-IF wireless FSK receiver. The IF is located at 3 MHz. implemented by delay-locked loop associated with the techniques of digital offset cancellation modified phase-frequency detection. demodulated data can be recovered one-clock latency. circuits in limiter RSSI are all pseudo differential to minimize requirement voltage headroom. Each gain cell feedforward common mode stabilization make sure its...