Updated output shapes for instance-segmentation models (#3407)

Author: Anna Grebneva

models/intel/instance-segmentation-person-0007/README.md

@@ -37,12 +37,14 @@ The expected channel order is `BGR`
 
 ## Outputs
 
-1. Name: `labels`, shape: `10` - Contiguous integer class ID for every
+Model has outputs with dynamic shapes.
+
+1. Name: `labels`, shape: `-1` - Contiguous integer class ID for every
    detected object.
-2. Name: `boxes`, shape: `10, 5` - Bounding boxes around every detected objects
+2. Name: `boxes`, shape: `-1, 5` - Bounding boxes around every detected objects
    in (top_left_x, top_left_y, bottom_right_x, bottom_right_y) format and its
    confidence score in range [0, 1].
-3. Name: `masks`, shape: `10, 224, 224` - Segmentation heatmaps for every output
+3. Name: `masks`, shape: `-1, 224, 224` - Segmentation heatmaps for every output
    bounding box.
 
 ## Training Pipeline

models/intel/instance-segmentation-security-0002/README.md

@@ -38,12 +38,14 @@ The expected channel order is `BGR`.
 
 ## Outputs
 
-1. Name: `labels`, shape: `100` - Contiguous integer class ID for every
+Model has outputs with dynamic shapes.
+
+1. Name: `labels`, shape: `-1` - Contiguous integer class ID for every
    detected object.
-2. Name: `boxes`, shape: `100, 5` - Bounding boxes around every detected objects
+2. Name: `boxes`, shape: `-1, 5` - Bounding boxes around every detected objects
    in (top_left_x, top_left_y, bottom_right_x, bottom_right_y) format and its
    confidence score in range [0, 1].
-3. Name: `masks`, shape: `100, 28, 28` - Segmentation heatmaps for every output
+3. Name: `masks`, shape: `-1, 28, 28` - Segmentation heatmaps for every output
    bounding box.
 
 ## Training Pipeline

models/intel/instance-segmentation-security-0091/README.md

@@ -38,12 +38,14 @@ The expected channel order is `BGR`
 
 ## Outputs
 
-1. Name: `labels`, shape: `100` - Contiguous integer class ID for every
+Model has outputs with dynamic shapes.
+
+1. Name: `labels`, shape: `-1` - Contiguous integer class ID for every
    detected object.
-2. Name: `boxes`, shape: `100, 5` - Bounding boxes around every detected objects
+2. Name: `boxes`, shape: `-1, 5` - Bounding boxes around every detected objects
    in (top_left_x, top_left_y, bottom_right_x, bottom_right_y) format and its
    confidence score in range [0, 1].
-3. Name: `masks`, shape: `100, 28, 28]` - Segmentation heatmaps for every output
+3. Name: `masks`, shape: `-1, 28, 28` - Segmentation heatmaps for every output
     bounding box.
 
 ## Training Pipeline

models/intel/instance-segmentation-security-0228/README.md

@@ -36,12 +36,14 @@ The expected channel order is `BGR`
 
 ## Outputs
 
-1. Name: `labels`, shape: `100` - Contiguous integer class ID for every
+Model has outputs with dynamic shapes.
+
+1. Name: `labels`, shape: `-1` - Contiguous integer class ID for every
    detected object.
-2. Name: `boxes`, shape: `100, 5` - Bounding boxes around every detected objects
+2. Name: `boxes`, shape: `-1, 5` - Bounding boxes around every detected objects
    in (top_left_x, top_left_y, bottom_right_x, bottom_right_y) format and its
    confidence score in range [0, 1].
-3. Name: `masks`, shape: `100, 14, 14` - Segmentation heatmaps for every output
+3. Name: `masks`, shape: `-1, 14, 14` - Segmentation heatmaps for every output
    bounding box.
 
 ## Training Pipeline

models/intel/instance-segmentation-security-1039/README.md

@@ -36,12 +36,14 @@ The expected channel order is `BGR`
 
 ## Outputs
 
-1. Name: `labels`, shape: `100` - Contiguous integer class ID for every
+Model has outputs with dynamic shapes.
+
+1. Name: `labels`, shape: `-1` - Contiguous integer class ID for every
    detected object.
-2. Name: `boxes`, shape: `100, 5` - Bounding boxes around every detected objects
+2. Name: `boxes`, shape: `-1, 5` - Bounding boxes around every detected objects
    in (top_left_x, top_left_y, bottom_right_x, bottom_right_y) format and its
    confidence score in range [0, 1].
-3. Name: `masks`, shape: `100, 14, 14` - Segmentation heatmaps for every output
+3. Name: `masks`, shape: `-1, 14, 14` - Segmentation heatmaps for every output
    bounding box.
 
 ## Training Pipeline

models/intel/instance-segmentation-security-1040/README.md

@@ -36,12 +36,14 @@ The expected channel order is `BGR`
 
 ## Outputs
 
-1. Name: `labels`, shape: `100` - Contiguous integer class ID for every
+Model has outputs with dynamic shapes.
+
+1. Name: `labels`, shape: `-1` - Contiguous integer class ID for every
    detected object.
-2. Name: `boxes`, shape: `100, 5` - Bounding boxes around every detected objects
+2. Name: `boxes`, shape: `-1, 5` - Bounding boxes around every detected objects
    in (top_left_x, top_left_y, bottom_right_x, bottom_right_y) format and its
    confidence score in range [0, 1].
-3. Name: `masks`, shape: `100, 28, 28` - Segmentation heatmaps for every output
+3. Name: `masks`, shape: `-1, 28, 28` - Segmentation heatmaps for every output
    bounding box.
 
 ## Training Pipeline