Vision Transformers Change The AI Acceleration Rules


Transformers were first introduced by the team at Google Brain in 2017 in their paper, "Attention is All You Need". Since their introduction, transformers have inspired a flurry of investment and research which have produced some of the most impactful model architectures and AI products to-date, including ChatGPT which is an acronym for Chat Generative Pre-trained Transformer. Transformers a... » read more

(Vision) Transformers: Rise Of The Chimera


It’s 2023 and transformers are having a moment. No, I’m not talking about the latest installment of the Transformers movie franchise, "Transformers: Rise of the Beasts"; I’m talking about the deep learning model architecture class, transformers, that is fueling anticipation, excitement, fear, and investment in AI. Transformers are not so new in the world of AI anymore; they were first ... » read more

Object Detection CNN Suitable For Edge Processors With Limited Memory


A technical paper titled “TinyissimoYOLO: A Quantized, Low-Memory Footprint, TinyML Object Detection Network for Low Power Microcontrollers” was published by researchers at ETH Zurich. Abstract: "This paper introduces a highly flexible, quantized, memory-efficient, and ultra-lightweight object detection network, called TinyissimoYOLO. It aims to enable object detection on microcontrol... » read more

Computational Imaging Craves System-Level Design And Simulation Tools To Leverage Disruptive AI In Embedded Vision


Image quality now relies more than ever on high computing power tied to miniaturized optics and sensors, rather than on standalone and bulky but aberration-free optics. This new trend is called computational imaging and can be used either for computational photography or for computer vision. Read this white paper to learn about market trends and promising system co-design and co-optimization ap... » read more

Machine Vision Plus AI/ML Adds Vast New Opportunities


Traditional technology companies and startups are racing to combine machine vision with AI/ML, enabling it to "see" far more than just pixel data from sensors, and opening up new opportunities across a wide swath of applications. In recent years, startups have been able to raise billions of dollars as new MV ideas come to light in markets ranging from transportation and manufacturing to heal... » read more

Performance Of Analog In-Memory Computing On Imaging Problems


A technical paper titled "Accelerating AI Using Next-Generation Hardware: Possibilities and Challenges With Analog In-Memory Computing" was published by researchers at Lund University and Ericsson Research. Abstract "Future generations of computing systems need to continue increasing processing speed and energy efficiency in order to meet the growing workload requirements under stringent en... » read more

Issues And Challenges In Super-Resolution Object Detection And Recognition


If you want high performance AI inference, such as Super-Resolution Object Detection and Recognition, in your SoC the challenge is to find a solution that can meet your needs and constraints. You need inference IP that can run the model you want at high accuracy. You need inference IP that can run the model at the frame rate you want: higher frame rate = lower latency, more time for dec... » read more

Working With The NimbleAI Project To Push The Boundaries Of Neuromorphic Vision


At the end of 2022, the EU kicked off a cool project that aims to implement neuromorphic vision. But what is that? Let’s take a deeper look at the project and our contribution. First, if you are not familiar with Codasip Labs, I want to mention this briefly. Codasip Labs is in fact our innovation hub where we explore new technologies and try to contribute to the technology of the future. ... » read more

Multiexpert Adversarial Regularization For Robust And Data-Efficient Deep Supervised Learning


Deep neural networks (DNNs) can achieve high accuracy when there is abundant training data that has the same distribution as the test data. In practical applications, data deficiency is often a concern. For classification tasks, the lack of enough labeled images in the training set often results in overfitting. Another issue is the mismatch between the training and the test domains, which resul... » read more

AI Feeds Vision Processor, Image Sensor Boom


Vision systems are rapidly becoming ubiquitous, driven by big improvements in image sensors as well as new types of sensors. While the sensor itself often is developed using mature-node silicon, increasingly it is connected to vision processors developed at the most advanced process nodes. That allows for the highest performance per watt, and it also allows designs to incorporate AI accelera... » read more

← Older posts