Live Updates: COVID-19 Cases
  • World 19,597,250
    World
    Confirmed: 19,597,250
    Active: 6,284,971
    Recovered: 12,587,296
    Death: 724,983
  • USA 5,098,494
    USA
    Confirmed: 5,098,494
    Active: 2,316,325
    Recovered: 2,618,025
    Death: 164,144
  • Brazil 2,967,064
    Brazil
    Confirmed: 2,967,064
    Active: 798,968
    Recovered: 2,068,394
    Death: 99,702
  • India 2,108,705
    India
    Confirmed: 2,108,705
    Active: 622,724
    Recovered: 1,443,183
    Death: 42,798
  • Russia 882,347
    Russia
    Confirmed: 882,347
    Active: 177,286
    Recovered: 690,207
    Death: 14,854
  • South Africa 545,476
    South Africa
    Confirmed: 545,476
    Active: 140,808
    Recovered: 394,759
    Death: 9,909
  • Mexico 469,407
    Mexico
    Confirmed: 469,407
    Active: 104,710
    Recovered: 313,386
    Death: 51,311
  • Peru 463,875
    Peru
    Confirmed: 463,875
    Active: 128,894
    Recovered: 314,332
    Death: 20,649
  • Chile 368,825
    Chile
    Confirmed: 368,825
    Active: 16,699
    Recovered: 342,168
    Death: 9,958
  • Spain 361,442
    Spain
    Confirmed: 361,442
    Active: 332,939
    Recovered: ?
    Death: 28,503
  • Iran 324,692
    Iran
    Confirmed: 324,692
    Active: 24,306
    Recovered: 282,122
    Death: 18,264
  • UK 309,005
    UK
    Confirmed: 309,005
    Active: 262,494
    Recovered: ?
    Death: 46,511
  • Saudi Arabia 287,262
    Saudi Arabia
    Confirmed: 287,262
    Active: 33,692
    Recovered: 250,440
    Death: 3,130
  • Pakistan 283,487
    Pakistan
    Confirmed: 283,487
    Active: 17,815
    Recovered: 259,604
    Death: 6,068
  • Bangladesh 255,113
    Bangladesh
    Confirmed: 255,113
    Active: 105,144
    Recovered: 146,604
    Death: 3,365
  • Italy 249,756
    Italy
    Confirmed: 249,756
    Active: 12,924
    Recovered: 201,642
    Death: 35,190
  • Turkey 238,450
    Turkey
    Confirmed: 238,450
    Active: 11,063
    Recovered: 221,574
    Death: 5,813
  • Germany 216,346
    Germany
    Confirmed: 216,346
    Active: 9,692
    Recovered: 197,400
    Death: 9,254
  • France 197,921
    France
    Confirmed: 197,921
    Active: 84,761
    Recovered: 82,836
    Death: 30,324
  • Canada 118,985
    Canada
    Confirmed: 118,985
    Active: 6,580
    Recovered: 103,435
    Death: 8,970
  • China 84,596
    China
    Confirmed: 84,596
    Active: 839
    Recovered: 79,123
    Death: 4,634
  • Netherlands 57,987
    Netherlands
    Confirmed: 57,987
    Active: 51,830
    Recovered: ?
    Death: 6,157
  • Australia 20,698
    Australia
    Confirmed: 20,698
    Active: 9,100
    Recovered: 11,320
    Death: 278
  • S. Korea 14,562
    S. Korea
    Confirmed: 14,562
    Active: 629
    Recovered: 13,629
    Death: 304
  • New Zealand 1,569
    New Zealand
    Confirmed: 1,569
    Active: 23
    Recovered: 1,524
    Death: 22

Google introduces convolutional network that has self-supervised tracking via video colorization

Author at TechGenyz Google
Google Colorizing Videos

Google takes a step further. Researchers at Google with the help of artificial intelligence introduces convolutional network that can not only add color to black and white videos but also can constrain those colors to particular objects, people, and pets in a given frame, like a child coloring within the lines of a flip book.

It has always been a problem for users to track objects in a video via computer vision. Visually tracking objects is challenging because it requires large, labeled tracking datasets for training, which are impractical to annotate at scale. And so Google makes use of a convolutional network that can put a color frame to the gray-scale videos.

We introduce a convolutional network that colorizes grayscale videos, but is constrained to copy colors from a single reference frame. – Carl Vondrick, Research Scientist, Machine Perception.

The Google colorizing videos, as the scientists describe is a convolutional neural network, a kind of neural network that is architecturally well-suited to object tracking and video stabilization. It helps to learn and to follow multiple objects through occlusions.

The first step taken was to teach the algorithm to color the gray-scale movies. Researchers taking clips from kinetics datasets like videos from YouTube that covers a wide range of human action converts the first frame to black and white. Then the neural network was trained by them to predict the original colors in subsequent frames.

Related

[mnky_related_posts num=”2″]

It was challenging to train the neural network as the model had to color moving objects and regions and was effectively forced to learn in order to track those objects and regions.

This forces the model to learn an explicit mechanism that we can use for tracking. – Vondrick

In the result, the model can keep tabs on any region specified in the first frame of the video and, if given points of reference, can even track human poses. It is greatly impressive to witness how it outperforms several state-of-the-art colorization techniques.

Source: Google AI Blog

Career

Subscribe