Updated input/output description of intel models (openvinotoolkit#3425)

Author: Anna Grebneva

models/intel/age-gender-recognition-retail-0013/README.md

@@ -38,7 +38,7 @@ applicable for children since their faces were not in the training set.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 62, 62` in `1, C, H, W` format, where:
+Image, name: `data`, shape: `1, 3, 62, 62` in `1, C, H, W` format, where:
 
 - `C` - number of channels
 - `H` - image height
@@ -48,7 +48,7 @@ Expected color order is `BGR`.
 
 ## Outputs
 
-1. Name: `age_conv3`, shape: `1, 1, 1, 1` - Estimated age divided by 100.
+1. Name: `fc3_a`, shape: `1, 1, 1, 1` - Estimated age divided by 100.
 2. Name: `prob`, shape: `1, 2, 1, 1` - Softmax output across 2 type classes [0 - female, 1 - male].
 
 ## Demo usage

models/intel/face-detection-0200/README.md

@@ -26,7 +26,7 @@ curve. All numbers were evaluated by taking into account only faces bigger than
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 256, 256` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 256, 256` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/face-detection-0202/README.md

@@ -26,7 +26,7 @@ curve. All numbers were evaluated by taking into account only faces bigger than
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 384, 384` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 384, 384` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/face-detection-0204/README.md

@@ -26,7 +26,7 @@ curve. All numbers were evaluated by taking into account only faces bigger than
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 448, 448` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 448, 448` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/face-detection-0205/README.md

@@ -25,7 +25,7 @@ curve. All numbers were evaluated by taking into account only faces bigger than
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 416, 416` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 416, 416` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/face-detection-0206/README.md

@@ -25,7 +25,7 @@ curve. All numbers were evaluated by taking into account only faces bigger than
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 640, 640` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 640, 640` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/face-detection-adas-0001/README.md

@@ -30,7 +30,7 @@ curve. Numbers are on
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 384, 672` in the format `B, C, H, W`, where:
+Image, name: `data`, shape: `1, 3, 384, 672` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/face-detection-retail-0004/README.md

@@ -27,7 +27,7 @@ curve. All numbers were evaluated by taking into account only faces bigger than
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 300, 300` in the format `B, C, H, W`, where:
+Image, name: `data`, shape: `1, 3, 300, 300` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/face-detection-retail-0005/README.md

@@ -26,7 +26,7 @@ curve. All numbers were evaluated by taking into account only faces bigger than
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 300, 300` in the format `B, C, H, W`, where:
+Image, name: `input.1`, shape: `1, 3, 300, 300` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/face-reidentification-retail-0095/README.md

@@ -34,7 +34,7 @@ To align the face, use a landmarks regression model: using regressed points and
 
 ## Inputs
 
-Input image, name: `data` , shape: `1, 3, 128, 128` in the format `B, C, H, W`, where:
+Input image, name: `0` , shape: `1, 3, 128, 128` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/faster-rcnn-resnet101-coco-sparse-60-0001/README.md

@@ -21,7 +21,7 @@ See Average Precision metric description at [COCO: Common Objects in Context](ht
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 800, 1280, 3` in the format `B, H, W, C`, where:
+Image, name: `image`, shape: `1, 800, 1280, 3` in the format `B, H, W, C`, where:
 
 - `B` - batch size
 - `H` - image height

models/intel/head-pose-estimation-adas-0001/README.md

@@ -46,9 +46,9 @@ Expected color order is `BGR`.
 Each output contains one float value that represents value in Tait-Bryan angles
 (yaw, pitch or roll).
 
-1. name: `angle_y_fc`, shape: `1, 1` - Estimated yaw (in degrees).
-2. name: `angle_p_fc`, shape: `1, 1` - Estimated pitch (in degrees).
-3. name: `angle_r_fc`, shape: `1, 1` - Estimated roll (in degrees).
+1. name: `fc_y`, shape: `1, 1` - Estimated yaw (in degrees).
+2. name: `fc_p`, shape: `1, 1` - Estimated pitch (in degrees).
+3. name: `fc_r`, shape: `1, 1` - Estimated roll (in degrees).
 
 ## Demo usage
 

models/intel/horizontal-text-detection-0001/README.md

@@ -21,7 +21,7 @@ The key benefit of this model compared to the [base model](../text-detection-000
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 704, 704` in the format `1, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 704, 704` in the format `1, C, H, W`, where:
 
 - `C` - number of channels
 - `H` - image height

models/intel/human-pose-estimation-0005/README.md

@@ -23,7 +23,7 @@ Average Precision metric described in [COCO Keypoint Evaluation site](https://co
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 288, 288` in the `B, C, H, W` format, where:
+Image, name: `image`, shape: `1, 3, 288, 288` in the `B, C, H, W` format, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/human-pose-estimation-0006/README.md

@@ -23,7 +23,7 @@ Average Precision metric described in [COCO Keypoint Evaluation site](https://co
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 352, 352` in the `B, C, H, W` format, where:
+Image, name: `image`, shape: `1, 3, 352, 352` in the `B, C, H, W` format, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/human-pose-estimation-0007/README.md

@@ -23,7 +23,7 @@ Average Precision metric described in [COCO Keypoint Evaluation site](https://co
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 448, 448` in the `B, C, H, W` format, where:
+Image, name: `image`, shape: `1, 3, 448, 448` in the `B, C, H, W` format, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/landmarks-regression-retail-0009/README.md

@@ -26,7 +26,7 @@ where N is the number of landmarks, _p_-hat and _p_ are, correspondingly, the pr
 
 ## Inputs
 
-Image, name: `data`, shape: `1, 3, 48, 48` in the format `B, C, H, W`, where:
+Image, name: `0`, shape: `1, 3, 48, 48` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels
@@ -37,7 +37,7 @@ The expected color order is `BGR`.
 
 ## Outputs
 
-The net outputs a blob with the shape: `1, 10`, containing a row-vector of 10 floating point values
+The net outputs a blob with the shape: `1, 10, 1, 1`, containing a row-vector of 10 floating point values
 for five landmarks coordinates in the form (x0, y0, x1, y1, ..., x4, y4).
 All the coordinates are normalized to be in range [0, 1].
 

models/intel/license-plate-recognition-barrier-0001/README.md

@@ -46,7 +46,7 @@ thoroughly. Other types of license plates may underperform.
 
 ## Outputs
 
-Encoded vector of floats, name: `decode`, shape: `1, 88, 1, 1`. Each float
+Encoded vector of floats, name: `dec`, shape: `1, 88, 1, 1`. Each float
 is an integer number encoding a character according to this dictionary:
 
         0 0

models/intel/pedestrian-and-vehicle-detector-adas-0001/README.md

@@ -27,7 +27,7 @@ Tested on challenging internal datasets with 1001 pedestrian and 12585 vehicles
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 384, 672` in the format `B, C, H, W`, where:
+Image, name: `data`, shape: `1, 3, 384, 672` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/pedestrian-detection-adas-0002/README.md

@@ -26,7 +26,7 @@ Tested on an internal dataset with 1001 pedestrian to detect.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 384, 672` in the format `B, C, H, W`, where:
+Image, name: `data`, shape: `1, 3, 384, 672` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/person-attributes-recognition-crossroad-0230/README.md

@@ -39,7 +39,7 @@ This model presents a person attributes classification algorithm analysis scenar
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 160, 80` in the format `1, C, H, W`, where:
+Image, name: `0`, shape: `1, 3, 160, 80` in the format `1, C, H, W`, where:
 
 - `C` - number of channels
 - `H` - image height

models/intel/person-detection-0106/README.md

@@ -24,7 +24,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 800, 1344` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 800, 1344` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels
@@ -35,15 +35,17 @@ Expected color order is `BGR`.
 
 ## Outputs
 
-1. The `boxes` is a blob with the shape `100, 5` in the format `N, 5`, where `N` is the number of detected
+Model has outputs with dynamic shapes.
+
+1. The `boxes` is a blob with the shape `-1, 5` in the format `N, 5`, where `N` is the number of detected
    bounding boxes. For each detection, the description has the format
    [`x_min`, `y_min`, `x_max`, `y_max`, `conf`], where:
 
     - (`x_min`, `y_min`) - coordinates of the top left bounding box corner
     - (`x_max`, `y_max`) - coordinates of the bottom right bounding box corner
     - `conf` - confidence for the predicted class
 
-2. The `labels` is a blob with the shape `100` in the format `N`, where `N` is the number of detected
+2. The `labels` is a blob with the shape `-1` in the format `N`, where `N` is the number of detected
    bounding boxes. It contains predicted class ID (0 - person) per each detected box.
 
 ## Demo usage

models/intel/person-detection-0200/README.md

@@ -25,7 +25,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 256, 256` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 256, 256` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/person-detection-0201/README.md

@@ -25,7 +25,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 384, 384` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 384, 384` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/person-detection-0202/README.md

@@ -25,7 +25,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 512, 512` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 512, 512` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/person-detection-0203/README.md

@@ -24,7 +24,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 480, 864` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 480, 864` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/person-detection-0301/README.md

@@ -24,7 +24,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 800, 1344` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 800, 1344` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/person-detection-0302/README.md

@@ -24,7 +24,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 720, 1280` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 720, 1280` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/person-detection-0303/README.md

@@ -24,7 +24,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 720, 1280` in the format `B, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 720, 1280` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels

models/intel/person-detection-action-recognition-0005/README.md

@@ -27,7 +27,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 400, 680` in the format `B, C, H, W`, where:
+Image, name: `data`, shape: `1, 3, 400, 680` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels
@@ -43,10 +43,10 @@ The net outputs four branches:
 1. name: `mbox_loc1/out/conv/flat`, shape: `b, num_priors*4` - Box coordinates in SSD format
 2. name: `mbox_main_conf/out/conv/flat/softmax/flat`, shape: `b, num_priors*2` - Detection confidences
 3. name: `mbox/priorbox`, shape: `1, 2, num_priors*4` - Prior boxes in SSD format
-4. name: `out/anchor1`, shape: `b, 3, h, w` - Action confidences
-5. name: `out/anchor2`, shape: `b, 3, h, w` - Action confidences
-6. name: `out/anchor3`, shape: `b, 3, h, w` - Action confidences
-7. name: `out/anchor4`, shape: `b, 3, h, w` - Action confidences
+4. name: `out/anchor1`, shape: `b, h, w, 3` - Action confidences
+5. name: `out/anchor2`, shape: `b, h, w, 3` - Action confidences
+6. name: `out/anchor3`, shape: `b, h, w, 3` - Action confidences
+7. name: `out/anchor4`, shape: `b, h, w, 3` - Action confidences
 
 Where:
 

models/intel/person-detection-action-recognition-teacher-0002/README.md

@@ -27,7 +27,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 400, 680` in the format `B, C, H, W`, where:
+Image, name: `data`, shape: `1, 3, 400, 680` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels
@@ -43,10 +43,10 @@ The net outputs four branches:
 1. name: `mbox_loc1/out/conv/flat`, shape: `b, num_priors*4` - Box coordinates in SSD format
 2. name: `mbox_main_conf/out/conv/flat/softmax/flat`, shape: `b, num_priors*2` - Detection confidences
 3. name: `mbox/priorbox`, shape: `1, 2, num_priors*4` - Prior boxes in SSD format
-4. name: `out/anchor1`, shape: `b, 3, h, w` - Action confidences
-5. name: `out/anchor2`, shape: `b, 3, h, w` - Action confidences
-6. name: `out/anchor3`, shape: `b, 3, h, w` - Action confidences
-7. name: `out/anchor4`, shape: `b, 3, h, w` - Action confidences
+4. name: `out/anchor1`, shape: `b, h, w, 3` - Action confidences
+5. name: `out/anchor2`, shape: `b, h, w, 3` - Action confidences
+6. name: `out/anchor3`, shape: `b, h, w, 3` - Action confidences
+7. name: `out/anchor4`, shape: `b, h, w, 3` - Action confidences
 
 Where:
 

models/intel/person-detection-asl-0001/README.md

@@ -22,7 +22,7 @@ Average Precision (AP) is defined as an area under the [precision/recall](https:
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 320, 320` in the format `1, C, H, W`, where:
+Image, name: `image`, shape: `1, 3, 320, 320` in the format `1, C, H, W`, where:
 
 - `C` - number of channels
 - `H` - image height

models/intel/person-detection-raisinghand-recognition-0001/README.md

@@ -27,7 +27,7 @@ curve.
 
 ## Inputs
 
-Image, name: `input`, shape: `1, 3, 400, 680` in the format `B, C, H, W`, where:
+Image, name: `data`, shape: `1, 3, 400, 680` in the format `B, C, H, W`, where:
 
 - `B` - batch size
 - `C` - number of channels
@@ -43,10 +43,10 @@ The net outputs four branches:
 1. name: `mbox_loc1/out/conv/flat`, shape: `b, num_priors*4` - Box coordinates in SSD format
 2. name: `mbox_main_conf/out/conv/flat/softmax/flat`, shape: `b, num_priors*2` - Detection confidences
 3. name: `mbox/priorbox`, shape: `1, 2, num_priors*4` - Prior boxes in SSD format
-4. name: `out/anchor1`, shape: `b, 2, h, w` - Action confidences
-5. name: `out/anchor2`, shape: `b, 2, h, w` - Action confidences
-6. name: `out/anchor3`, shape: `b, 2, h, w` - Action confidences
-7. name: `out/anchor4`, shape: `b, 2, h, w` - Action confidences
+4. name: `out/anchor1`, shape: `b, h, w, 2` - Action confidences
+5. name: `out/anchor2`, shape: `b, h, w, 2` - Action confidences
+6. name: `out/anchor3`, shape: `b, h, w, 2` - Action confidences
+7. name: `out/anchor4`, shape: `b, h, w, 2` - Action confidences
 
 Where:
 

models/intel/person-detection-retail-0002/README.md

@@ -36,7 +36,7 @@ curve. Validation dataset consists of ~50K of images from ~100 different scenes.
 
     The expected channel order is `BGR`.
 
-2. name: `im_info`, shape: `1x6` - An image information
+2. name: `im_info`, shape: `1, 6` - An image information
     [544, 992, 992/`frame_width`, 544/`frame_height`, 992/`frame_width`, 544/`frame_height`]
 
 ## Outputs