iSIZE is a deep-tech company that specializes in deep learning for efficient, intelligent and sustainable video delivery. iSIZE patented technology is powered by the latest AI innovations to enable clients to reach more users, provide better experiences and reduce the financial and environmental costs associated with video delivery. iSIZE products provide innovations spanning encoding, denoising and generative video content creation. They reduce latency, enhance video quality, and integrate easily with any existing codec by using a deep neural network that improves itself over time, without breaking any standards. This unique approach results in substantial bandwidth, energy and cost savings for video entertainment platforms, gaming, VR/ AR, IoT, VoD and live streaming services.
More information is available from www.isize.co
Products and Services
iSIZE has developed a deep perceptual ‘precoder’ – a software solution that uses AI-trained to “see with the human eye” in order to optimize visual quality in order to save video bitrate during encoding.
Our flagship product, BitSave is available both as a SaaS platform at bitsave.tech and for on-premise use (currently available on AWS Marketplace and in the future on Microsoft Azure, Google Cloud and other cloud platforms). BitSave is an AI-based perceptual preprocessing solution that allows conventional, third-party encoders to produce higher quality video at lower bitrate. BitSave achieves this by preprocessing the video content before it reaches the video encoder, such that the output after compressing with any standard video codec is perceptually optimized with less motion artifacts or blurring, for the same or lower bitrate.
BitSave is able to reduce bitrate requirements for a given quality level by up to 30% versus that encoder. When coupled with open-source AVC, HEVC and AV1 encoders, BitSave allows for up to 40% bitrate reduction at no quality compromise in comparison to leading third-party AVC, HEVC and AV1 encoding services. Lower bitrates equate to smaller file sizes, lower storage, transmission and distribution costs, reduced energy consumption and more satisfied customers.
This results in substantial bandwidth, energy and cost savings for VoD and live streaming services, broadcasters, cloud gaming, VR/AR, IoT and the end consumers.
- BitSave Solution Brief
- BitSave Summary
- BitSave Datasheet
- BitSave Demo
- BitSave wins 2021 VideoTech Innovation Award for Sustainability
BitClear is an AI-based video processing technology that removes compression artifacts (like blurring and blocking artifacts) from user-generated (or heavily-compressed) content. BitClear provides a scalable solution to denoise and upscale severely distorted low-resolution video assets.
Video content that has been through multiple transcoding iterations can be revived to the maximum possible quality without affecting the artistic intent of the original creators. The process can also allow for video upscaling, all with as little as 5ms processing latency on GPUs or high-performance CPUs. BitClear is 2 to 10 times faster than the state-of-the-art in video denoising and upscaling, while offering superior quality.
BitClear is designed to achieve the maximum quality improvement by alleviating the effect of compression artifacts. Its AI nature and implementation efficiency means it can be deployed at scale without the need for human inspection or tuning. This results in exceptional video quality and real time denoising for social media, UGC distribution, streaming companies, live game streams, live sports, entertainment, and virtual reality feeds. BitClear can also be used for IoT and Defence applications by recovering image quality of low bandwidth feeds - see more in the links below.
- BitClear Solution Brief
- BitClear Summary
- BitClear Datasheet
- BitClear Demo
- BitClear Press Release
- BitClear wins Product of the Year at NAB 2022
- BitClear Defence Applications
- BitClear wins Best of Show at IBC 2022
More detailed information about our technology can be found in iSIZE Blogs and in iSIZE Technical Papers and Presentations
- AI Driven Live Denoising and Upscaling up to 4K with Intel Advanced Matrix Extensions
- Intel Network Builders Winners' Circle Gold Award
- Innovator at the Edge - Intel Tech Talk
- iSIZE at ACM Multimedia 2022, presentation with Meta
- Creating Perceptually Optimized Videos In The Cloud
- Reverse-Engineering Visual Quality Through Pixel-to-Pixel Deep Neural Networks
iSIZE Technical Papers and Presentations
- Domain-specific fusion of multiple objective quality metrics - VQEG 2022 | May 9-13, 2022 (Press release)
- Towards generalized psychovisual preprocessing for video encoding - SMPTE Motion Imagine Journal | May 10, 2022 (Press release)
- Leveraging AI to improve video quality - NAB 2022 | April 23-27, 2022
- Towards generalized psychovisual preprocessing for video encoding - SMPTE 2021 | November 7, 2021 (Press release)
- Deep perceptual preprocessing for video coding - CVPR 2021 | 19 June, 2021 (Press release)
- Escaping the complexity-bitrate-quality barriers of video encoders via deep perceptual optimization - SPIE Optical Engineering and Aplications | August 21, 2020
- IBC2019 Conference - Tech Talk: AI - more ways it will revolutionise our industry - Min 21:10 | September 17, 2019