no code implementations • 15 Dec 2023 • Yu-Han Sun, Chiang Lo-Hsuan Lee, Tian-Sheuan Chang
Image prefiltering with just noticeable distortion (JND) improves coding efficiency in a visual lossless way by filtering the perceptually redundant information prior to compression.
no code implementations • 13 Dec 2023 • Yuan YAO, Tian-Sheuan Chang
Furthermore, the hardware architecture scales effectively, with only a sublinear increase in area cost.
no code implementations • 30 Aug 2023 • Tun-Hao Yang, Tian-Sheuan Chang
This challenge leads many accelerators to opt for simpler and shallow models like FSRCNN, compromising performance for real-time needs, especially for resource-limited edge devices.
no code implementations • 10 May 2022 • Jien-De Sui, Wei-Han Chen, Tzyy-Yuang Shiang, Tian-Sheuan Chang
Previous gait phase detection as convolutional neural network (CNN) based classification task requires cumbersome manual setting of time delay or heavy overlapped sliding windows to accurately classify each phase under different test cases, which is not suitable for streaming Inertial-Measurement-Unit (IMU) sensor data and fails to adapt to different scenarios.
no code implementations • 10 May 2022 • Yu-Hsiang Chiang, Tian-Sheuan Chang, Shyh Jye Jou
Keyword spotting has gained popularity as a natural way to interact with consumer devices in recent years.
no code implementations • 9 May 2022 • An-Jung Huang, Kai-Chieh Hsu, Tian-Sheuan Chang
Deep learning based superresolution achieves high-quality results, but its heavy computational workload, large buffer, and high external memory bandwidth inhibit its usage in mobile devices.
no code implementations • 9 May 2022 • Hong-Yi Wang, Tian-Sheuan Chang
Following the success of the natural language processing, the transformer for vision applications has attracted significant attention in recent years due to its excellent performance.
no code implementations • 9 May 2022 • Yu-Hsiang Chiang, Cheng En Ni, Yun Sung, Tuo-Hung Hou, Tian-Sheuan Chang, Shyh Jye Jou
However, in-RRAM computing (IRC) suffered from large device variation and numerous nonideal effects in hardware.
no code implementations • 6 May 2022 • Jien-De Sui, Tian-Sheuan Chang
The proposed model can achieve better average percent error, 4. 78\%, on running and walking stride length regression and 99. 83\% accuracy on running and walking classification, when compared to the previous approach, 7. 44\% on the stride length estimation.
no code implementations • 2 May 2022 • Hsu-Tung Shih, Tian-Sheuan Chang
The large amount of memory bandwidth between local buffer and external DRAM has become the speedup bottleneck of CNN hardware accelerators, especially for activation maps.
no code implementations • 2 May 2022 • Yi-An Chen, Jien-De Sui, Tian-Sheuan Chang
Gait phase detection with convolution neural network provides accurate classification but demands high computational cost, which inhibits real time low power on-sensor processing.
no code implementations • 2 May 2022 • Dun-Hao Yang, Tian-Sheuan Chang
Increasingly, convolution neural network (CNN) based super resolution models have been proposed for better reconstruction results, but their large model size and complicated structure inhibit their real-time hardware implementation.
no code implementations • 2 May 2022 • Hong-Han Lien, Tian-Sheuan Chang
Spiking neural networks (SNNs), which are inspired by the human brain, have recently gained popularity due to their relatively simple and low-power hardware for transmitting binary spikes and highly sparse activation maps.
no code implementations • 2 May 2022 • Shu-Hung Kuo, Tian-Sheuan Chang
Computing-in-memory (CIM) has attracted significant attentions in recent years due to its massive parallelism and low power consumption.
no code implementations • 2 May 2022 • Chih-Chyau Yang, Tian-Sheuan Chang
With the popularity of the deep neural network (DNN), hardware accelerators are demanded for real time execution.
no code implementations • 2 May 2022 • Kuo-Wei Chang, Hsu-Tung Shih, Tian-Sheuan Chang, Shang-Hong Tsai, Chih-Chyau Yang, Chien-Ming Wu, Chun-Ming Huang
Memory bandwidth has become the real-time bottleneck of current deep learning accelerators (DLA), particularly for high definition (HD) object detection.
no code implementations • 2 May 2022 • Kuo-Wei Chang, Tian-Sheuan Chang
Hardware acceleration for dilated and transposed convolution enables real time execution of related tasks like segmentation, but current designs are specific for these convolutional types or suffer from complex control for reconfigurable designs.
no code implementations • 16 Dec 2017 • Chih-Cheng Chang, Pin-Chun Chen, Teyuh Chou, I-Ting Wang, Boris Hudec, Che-Chia Chang, Chia-Ming Tsai, Tian-Sheuan Chang, Tuo-Hung Hou
Asymmetric nonlinear weight update is considered as one of the major obstacles for realizing hardware neural networks based on analog resistive synapses because it significantly compromises the online training capability.