90 likes | 111 Views
Video streaming is predicted to account for a staggering 82 percent of the total internet traffic by 2022. From basic android mobile to 4K premium tv u2013 viewers are watching content on devices with varied resolutions, compute & capability. The massive amount of data streaming presents challenges to broadcasters, content distribution networks, such as buffering issues, low resolution, poor quality, high operating cost, monetization, etc. AI/ML helps to overcome these challenges by providing the highest possible quality of experience (QoE) and quality of service (QoS).<br>AI/ML models can enable broadcasters and content providers to unearth opportunities to reduce streaming costs, enhance the viewer quality of experience, and improve viewer engagement using smart advertisement & video analytics by using AI/ML models throughout the media workflow.<br>This blog discusses the use of different AI/ML models in video streaming to enhance QoE and QoS.<br>
E N D
Welcome to PowerPoint 5 tips for a simpler way to work AI/ML FOR / BROADCAST / VIDEO STREAMING
Reduce cost and Monetize Video Streaming with AI/ML • Video streaming is predicted to account for a staggering 82 percent of the total internet traffic by 2022. From basic android mobile to 4K premium tv – viewers are watching content on devices with varied resolutions, compute & capability. The massive amount of data streaming presents challenges to broadcasters, content distribution networks, such as buffering issues, low resolution, poor quality, high operating cost, monetization, etc. AI/ML helps to overcome these challenges by providing the highest possible quality of experience (QoE) and quality of service (QoS). • AI/ML models can enable broadcasters and content providers to unearth opportunities to reduce streaming costs, enhance the viewer quality of experience, and improve viewer engagement using smart advertisement & video analytics by using AI/ML models throughout the media workflow.
Introduction AI/ML in Video Streaming Video Streaming Devices with AI/ML and Neural Network models will surely outperform the ones without by • Enhanced streaming quality • Intelligent video experience • Ease of use Gyrus using its various AI/ML and Neural Networks models provide solutions for • Improve the quality of a image maintaining the resolution • UP Scaling Low res to High resolution • Removing watermarks & damages from the Images • Frame rate conversion and Integrated Home Assistant in TV • Person detection, Noise reduction, Automatic scene detection and Voice controls Transformation in Video Streaming Devices
AI/ML Algorithm to provide Optimal settings to reduce cost • Inputs • Device specific information • Display Size • Codecs’ available • User Information • Cost Functions • Network / Speed • Storage • Compute Costs • Content Information • Content Type Optionally Auto Classified • Source Stream Options • Optimization Parameters • Cost / Viewing • Output • Transcoding Parameters • Data Rate per stream • Post Processing Options
Automatic Video Synthesis Top left - Original Labels, Top right - Original Output, Bottom left - Buildings to Trees, Bottom right - Trees to Buildings • Banners / Billboards with custom Ads • In Scene object replacement • Texture Changes per viewer preference
Super Resolution using Neural Networks 1 2 3 4 Super-resolution is a technique that enhances the quality of a given low- resolution to high resolution by upscaling. Upscaling Neural Network Models by GAN & CONV/ DECONV based. Features & Advantages Produces resolutions that are 2-4 times higher than the pixel count of the sensor Upscaling better than the traditional bilinear scaling. Super Resolution with and without Neural Networks
Frame Rate Conversion 1. Neural networks will be out-performing traditional methods in frame rate conversion and up-sampling. 2. The image above shows how using Variational Autoencoder, GANs, and Deep Convolution Networks can be used to perform this function and Deep Convolution Networks seem to have performed better than other and traditional upsampling. 3. Interpolation and upscaling for animated graphics is hard as it has sharper edges and simple interpolation tends to make the images blurred. Frame Rate Conversion example
Automatic Scene Detection • Every frame can be classified for detection of Specific Objects/location, Face recognition of the character and Auto Classification of Nudity. • There are several applications that can use the information about the running image / scene. • The insights can be used for auto captions for regulatory warnings such as Smoking / Drinking / etc Dividing a video into scenes and using AI to label these segments enables the creation of an inverted index for video content that is searchable.
Conclusion • We are just at the beginning of the applications of AI to video streaming. Identifying objects, recognizing faces, inserting customized advertisements, complying with regulatory warnings, generating subtitles at high speeds, transmitting the video intelligently at low bandwidth and lower costs are a few of the tasks that can be managed by AI engines very effectively. AI and deep neural network-based enhancements will dramatically improve user QoE and alter video consumption forever. AI solutions will soon become a widespread standard and will continue to redefine the streaming media ecosystem with emerging