ImageContainsQ
HighlightImage
ImageBoundingBoxes
ImagePosition
VideoMapTimeSeries
VideoFrameList
VideoFrameMap
ImageAssemble
ImageAdd
TimeSeriesMap

Image Processing
Computer Vision
Video Processing

Object Recognition & Tracking in Videos

Detect and track an object in a video

Open in Cloud
Download Notebook

Download the source video:

In[1]:=

v=ResourceData["Example""Object Recognition & Tracking in Videos","Video"]

Out[1]=

Use

ImageContainsQ

to find whether a dog is detected in each frame:

In[2]:=

check=VideoMapTimeSeriesImageContainsQ#Image,

domestic dog

CONCEPT

&,v;

Create a plot showing in which frames of the video a dog is detected:

In[3]:=

ListPlotTimeSeriesMap[Boole,check],

options



Out[3]=

Alternatively, extract the video frames and assemble a visual representation of the marked frames:

In[4]:=

images=VideoFrameList[Video[v,RasterSize200],All];

In[5]:=

ImageAssemble[Partition[MapThread[ImageAdd[ImageResize[#,25],#2{0,.4,0}]&,{images,Boole[Values[check]]}],15]]

Out[5]=

Using

ImagePosition

, you can track the movement of the dog:

In[6]:=

pos=ImagePosition#,

domestic dog

CONCEPT

&/@images;

This is the trajectory detected in the video, over time:

In[7]:=

HighlightImageVideoExtractFramesVideov,

...

,

,Line[Flatten[pos,1]]

Out[7]=

Overlay the detected bounding box on the original video frame:

In[8]:=

VideoFrameMapHighlightImage#,ImageBoundingBoxes#,

domestic dog

CONCEPT

&&,v

Out[8]=

Related Symbols

ImageContainsQ
HighlightImage
ImageBoundingBoxes
ImagePosition
VideoMapTimeSeries
VideoFrameList
VideoFrameMap
ImageAssemble
ImageAdd
TimeSeriesMap

Publisher Information

Contributed by: Wolfram Staff

License Information

This work is licensed under a Creative Commons Attribution 4.0 International License

Send a message about this example
Top

Wolfram Language Example Repository

Object Recognition & Tracking in Videos

See Also

Publisher Information

License Information