serve/*.c

/bbf/detect.objects

Using BBF classifier cascade to detect objects in a given image.

• ‘source’ or HTTP body : the image.
• ‘model’ : what object, for now, it only supports ‘face’.

You can look up the rest of parameters at ccv_bbf.c.

Supported methods: GET, POST

/convnet/classify

Using ConvNet to categorize a given image.

• ‘source’ or HTTP body : the image.
• ‘model’ : what category set, for now it only supports ‘image-net’.
• ‘top’ : the number of results returned, order by confidence score.

Supported methods: GET, POST

/dpm/detect.objects

Using DPM mixture model to detect objects in a given image.

• ‘source’ or HTTP body : the image.
• ‘model’ : what object, it now supports ‘pedestrian’ and ‘car’.

You can look up the rest of parameters at ccv_dpm.c.

Supported methods: GET, POST

/icf/detect.objects

Using ICF classifier cascade to detect objects in a given image.

• ‘source’ or HTTP body : the image.
• ‘model’ : what object, for now, it only supports ‘pedestrian’.

You can look up the rest of parameters at ccv_icf.c.

Supported methods: GET, POST

/swt/detect.words

Using SWT to detect words / texts in a given image.

• ‘source’ or HTTP body : the image.

You can look up the rest of parameters at ccv_swt.c.

Supported methods: GET, POST

/tld/track.object

Create a new TLD tracking instance with the initial frame.

• ‘source’ or HTTP body : the initial frame image.
• ‘x’ : the initial tracking rectangle’s top left coordinate.
• ‘y’ : the initial tracking rectangle’s top left coordinate.
• ‘width’ : the initial tracking rectangle’s width.
• ‘height’ : the initial tracking rectangle’s height.

You can look up the rest of parameters at ccv_tld.c.

Supported methods: GET, POST

On success, it will return the new tracking instance with ‘Location’ header, you can also find its ID in response[‘tld’].

/tld/track.object/[\d+]

Continue a TLD tracking instance with follow up frames.

• ‘source’ or HTTP body : the next frame image.
• ‘previous’ : the previous frame image, please make sure this is the exact copy of the frame you previous provided, otherwise API will return ‘false’.

Supported methods: GET, POST, DELETE

Please make sure that you DELETE the TLD tracking instance once you are done, otherwise the HTTP server cannot reclaim the memory it occupies.

/sift

Run SIFT feature point extraction on a given image.

• ‘source’ or HTTP body : the image.

You can look up the rest of parameters at ccv_sift.c.

Supported methods: GET, POST

‹ back

comments powered by Disqus