image file in bytes
clear ImageFeature
copy the float array to a storage
copy the float array to a storage
destination array
offset to copy
key that maps float array
BGR to RGB
image pixels in float array
image pixels in float array
key that maps float array
float array
get current channel
get current height
imInfo is a tensor that contains height, width, scaleInHeight, scaleInWidth e.
imInfo is a tensor that contains height, width, scaleInHeight, scaleInWidth e.g. it is used in SSD and Faster-RCNN to post process the roi detection
get label from ImageFeature
get original height
get original image size in (height, width, channel)
get original image size in (height, width, channel)
(height, width, channel)
get original width
get current image size in (height, width, channel)
get current image size in (height, width, channel)
(height, width, channel)
get current width
whether this ImageFeature contains label
whether this ImageFeature contains label
whether this image feature is valid
get opencv mat from ImageFeature, note that it may be empty if it is released
get prediction result from ImageFeature
get prediction result from ImageFeature
key that maps prediction result
Convert ImageFeature to image tensor
Convert ImageFeature to image tensor
key that maps the float array
transpose the image from hwc to chw
tensor that represents an image
get uri from ImageFeature
get uri from ImageFeature
Each ImageFeature keeps information about single image, it can include various status of an image, e.g. original bytes read from image file, an opencv mat, pixels in float array, image label, meta data and so on. it uses HashMap to store all these data, the key is string that identify the corresponding value