dharunselvaraju91

1
Impact of Artificial Intelligence Machine
Learning In Video Streaming Devices

• Augmenting TV user Experience

2
Introduction

• AI/ML in Video Streaming Devices
• 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
3
Super Resolution using Neural Networks
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
4
InPainting with Neural Networks

• How Inpainting with Neural Networks works
• Remove the watermarks
• Repair the damages
• Deconv and GAN Models are used
• The single network can be trained to do both
  Super Resolution and Inpainting (and other noise
  reduction) together.

InPainting with Neural Networks
5
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
6
Anonymized Person Detection
New AI techniques are available to anonymize
incoming video streaming at the source and
perform person recognition at the ball/stick level
Neural Networks identifies the user based on the
gait of the person without intruding into privacy.
Above all functions performed without intruding
into the privacy of the users and without
recording specific users video/face etc.
Person detect from Ball/Stick models
7
Integrated Home Assistant in TV

• AI integrated Video streaming devices at home can
  keep track of each individuals preferences,
  viewing habits and specific programming.
• Seven in 10 homes now have an SVOD service and
  72 use video streaming-capable TV devices.
• A background application running on TV can
  perform the task of recommending the specific
  content as per individuals preferences and
  viewing habits.

8
Noise Reduction

• Post-processing can remove noise and improve PSNR
  caused by Encoders blocking, Blurring, Ringing
  and Perceptual Quantizer (PQ).
• Using Neural Networks can improve the performance
  of noise reduction and they all can be performed
  by one algorithm.
• Post-processing options in TVs to perform MEMC
  can be enhanced with Neural Networks.

9
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.
10
Voice Controls

• Very similar to the Home AI assistants such as
  Amazon Alexa and Google Home, Video streaming
  devices at home expected to respond to voice
  commands
• The Wake-word detection and Automatic Speech
  Recognition (ASR) are implemented using complex
  Neural Network today.

Gyrus ML/AI algorithms can be used to fuse the
data and also to derive all the insights
presented above.