roi

Type Members

1. class BatchSampler extends Serializable

   

   sample box from given parameters, and regard it as positive if it satisfies overlap constraints

2. class RandomSampler extends Crop

   

   Random sample a bounding box given some constraints and crop the image This is used in SSD training augmentation

3. case class RoiHFlip(normalized: Boolean = true) extends FeatureTransformer with Product with Serializable

   

   horizontally flip the roi

   horizontally flip the roi

   normalized

   whether the roi is normalized, i.e. in range [0, 1]

4. case class RoiLabel(classes: Tensor[Float], bboxes: Tensor[Float], masks: Array[SegmentationMasks] = null) extends Product with Serializable

   

   image target with classes and bounding boxes

   image target with classes and bounding boxes

   classes

   N (class labels) or 2 * N, the first row is class labels, the second line is difficults

   bboxes

   N * 4, (xmin, ymin, xmax, ymax)

   masks

   the array of annotation masks of the targets

5. case class RoiNormalize() extends FeatureTransformer with Product with Serializable

   

   Normalize Roi to [0, 1]

6. case class RoiProject(needMeetCenterConstraint: Boolean = true) extends FeatureTransformer with Product with Serializable

   

   Project gt boxes onto the coordinate system defined by image boundary

   Project gt boxes onto the coordinate system defined by image boundary

   needMeetCenterConstraint

   whether need to meet center constraint, i.e., the center of gt box need be within image boundary

7. case class RoiResize(normalized: Boolean = false) extends FeatureTransformer with Product with Serializable

   

   resize the roi according to scale

   resize the roi according to scale

   normalized

   whether the roi is normalized, i.e. in range [0, 1]