Related papers: A Reconfigurable Convolution-in-Pixel CMOS Image S…
Transformer-based models are becoming more and more intelligent and are revolutionizing a wide range of human tasks. To support their deployment, AI labs offer inference services that consume hundreds of GWh of energy annually and charge…
Contrastive Vision-Language Pre-training, known as CLIP, has provided a new paradigm for learning visual representations by using large-scale contrastive image-text pairs. It shows impressive performance on zero-shot knowledge transfer to…
Dilated Convolutions have been shown to be highly useful for the task of image segmentation. By introducing gaps into convolutional filters, they enable the use of larger receptive fields without increasing the original kernel size. Even…
Single-pixel imaging (SPI) is an emerging technique which has attracts wide attention in various research fields. However, restricted by the low reconstruction quality and large amount of measurements, the practical application is still in…
Perceptual image compression has shown strong potential for producing visually appealing results at low bitrates, surpassing classical standards and pixel-wise distortion-oriented neural methods. However, existing methods typically improve…
Generative model based image lossless compression algorithms have seen a great success in improving compression ratio. However, the throughput for most of them is less than 1 MB/s even with the most advanced AI accelerated chips, preventing…
The performance and efficiency of running large-scale datasets on traditional computing systems exhibit critical bottlenecks due to the existing "power wall" and "memory wall" problems. To resolve those problems, processing-in-memory (PIM)…
To understand the scientific imaging capability, one must characterize the intra-pixel sensitivity variation (IPSV) of the CMOS image sensor. Extracting an IPSV map contributes to an improved detector calibration that allows to eliminate…
The future of artificial intelligence (AI) acceleration demands a paradigm shift beyond the limitations of purely electronic or photonic architectures. Photonic analog computing delivers unmatched speed and parallelism but struggles with…
Recent advances in photonic inverse design have demonstrated the ability to automatically synthesize compact, high-performance photonic components that surpass conventional, hand-designed structures, offering a promising path toward…
Data movement between memory and processors is a major bottleneck in modern computing systems. The processing-in-memory (PIM) paradigm aims to alleviate this bottleneck by performing computation inside memory chips. Real PIM hardware (e.g.,…
Computing-In-Memory (CIM) offers a potential solution to the memory wall issue and can achieve high energy efficiency by minimizing data movement, making it a promising architecture for edge AI devices. Lightweight models like MobileNet and…
The massive amounts of data generated by camera sensors motivate data processing inside pixel arrays, i.e., at the extreme-edge. Several critical developments have fueled recent interest in the processing-in-pixel-in-memory paradigm for a…
Recent progress on Transformers and multi-layer perceptron (MLP) models provide new network architectural designs for computer vision tasks. Although these models proved to be effective in many vision tasks such as image recognition, there…
In this paper, we present a novel transformer-based architecture for end-to-end image compression. Our architecture incorporates blocks that effectively capture local dependencies between tokens, eliminating the need for positional encoding…
Digital processing-in-memory (PIM) architectures mitigate the memory wall problem by facilitating parallel bitwise operations directly within the memory. Recent works have demonstrated their algorithmic potential for accelerating…
Tractable models of human perception have proved to be challenging to build. Hand-designed models such as MS-SSIM remain popular predictors of human image quality judgements due to their simplicity and speed. Recent modern deep learning…
We propose an end-to-end image compression and analysis model with Transformers, targeting to the cloud-based image classification application. Instead of placing an existing Transformer-based image classification model directly after an…
CMOS sensors were successfully implemented in the STAR tracker [1]. LHC experiments have shown that efficient b tagging, reconstruction of displaced vertices and identification of disappearing tracks are necessary. An improved vertex…
In this paper, an in-pixel chopping technique to reduce the low-frequency or 1/f noise of the source follower (SF) transistor in an active pixel sensor (APS) is presented. The SF low-frequency noise is modulated at higher frequencies…