Mask RCNN 實戰(一)－－程式碼詳細解析

阿新 • • 發佈：2019-02-01

Mask RCNN:（大家有疑問的請在評論區留言）

如果對原理不瞭解的話，可以花十分鐘先看一下我的這篇博文，在來進行實戰演練，這篇博文將是讓大家對mask rcnn 進行一個入門，我在後面的博文中會介紹mask rcnn 如何用於多人關鍵點檢測和多人姿態估計，以及如何利用mask rcnn 訓練自己的資料集，以及mobile_net版的mask rcnn（ps:我正在做，等做完我會分享到我的github上，感興趣的朋友可以繼續關注我後續的部落格，很快就會更新。）

大家先到GitHub上下載專案原始碼：Mask RCNN專案原始碼

一、配置環境：

Mask R-CNN是基於Python3，Keras，TensorFlow

。

Python 3.4+
TensorFlow 1.3+
Keras 2.0.8+
Jupyter Notebook
Numpy, skimage, scipy, Pillow, cython, h5py

二、安裝

安裝依賴關係
```
  pip3 install -r requirements.txt 
```
克隆這個儲存庫
從儲存庫根目錄執行安裝程式
```
  python3 setup.py install
```

如果需要在COCO資料集上訓練或測試，需要安裝pycocotools， clone下來，make生成對應的檔案，make之後將生成的pycocotools資料夾複製到samples中的coco資料夾。

coco資料集下載：

要在MS COCO上進行培訓或測試，您還需要：

pycocotools（安裝說明如上）
MS COCO資料集(Ubuntu 建議採用 wget 命令直接Ubuntu終端下載)

wget http://images.cocodataset.org/zips/train2014.zip #下載coco train2014訓練集圖片

wget http://images.cocodataset.org/zips/val2014.zip #下載coco val2014驗證集圖片

wget http://images.cocodataset.org/zips/test2014.zip #下載coco test2014 測試集圖片

wget https://dl.dropboxusercontent.com/s/o43o90bna78omob/instances_minival2014.json.zip?dl=0 #下載minival

wget https://dl.dropboxusercontent.com/s/s3tw5zcg7395368/instances_valminusminival2014.json.zip?dl=0 #下載validation-minus-minival

如果您使用Docker，則程式碼已驗證可在此Docker容器上使用（如果下載不下來，請在評論區留郵箱）。

它展示了一個使用MS COCO預先訓練的模型來分割自己影象中的物件的例子。它包括在任意影象上執行物件檢測和例項分割的程式碼。
這款筆記本引入了一個玩具資料集（Shapes）來演示新資料集的訓練。
它提供了管道每一步的視覺化。

1、開啟samples裡面demo.ipynb（用jupyter notebook）

demo.ipynb

import os
import sys
import random
import math
import numpy as np
import skimage.io
import matplotlib
import matplotlib.pyplot as plt

# Root directory of the project
ROOT_DIR = os.path.abspath("../Desktop/Mask_RCNN-master") #專案的資料夾

# Import Mask RCNN
sys.path.append(ROOT_DIR)  # To find local version of the library
import mrcnn.utils
import mrcnn.model as modellib
from mrcnn import visualize
# Import COCO config
sys.path.append(os.path.join(ROOT_DIR, "samples/coco/"))  # 匯入coco資料集，即下載5Kminival和35K validataon-minus-minival子集 （放入coco資料夾中）
import coco

%matplotlib inline 

# Directory to save logs and trained model
MODEL_DIR = os.path.join(ROOT_DIR, "logs")

# Local path to trained weights file
COCO_MODEL_PATH = os.path.join(ROOT_DIR, "mask_rcnn_coco.h5") #下載的訓練權重放入專案資料夾中
# Download COCO trained weights from Releases if needed
if not os.path.exists(COCO_MODEL_PATH):
    utils.download_trained_weights(COCO_MODEL_PATH)

# Directory of images to run detection on
IMAGE_DIR = os.path.join(ROOT_DIR, "images")  #測試圖片

Configurations

class InferenceConfig(coco.CocoConfig):
    # Set batch size to 1 since we'll be running inference on
    # one image at a time. Batch size = GPU_COUNT * IMAGES_PER_GPU
    GPU_COUNT = 1
    IMAGES_PER_GPU = 1
config = InferenceConfig()
config.display()

Configurations:
BACKBONE_SHAPES                [[256 256]
 [128 128]
 [ 64  64]
 [ 32  32]
 [ 16  16]]
BACKBONE_STRIDES               [4, 8, 16, 32, 64]
BATCH_SIZE                     1
BBOX_STD_DEV                   [ 0.1  0.1  0.2  0.2]
DETECTION_MAX_INSTANCES        100
DETECTION_MIN_CONFIDENCE       0.5
DETECTION_NMS_THRESHOLD        0.3
GPU_COUNT                      1
IMAGES_PER_GPU                 1
IMAGE_MAX_DIM                  1024
IMAGE_MIN_DIM                  800
IMAGE_PADDING                  True
IMAGE_SHAPE                    [1024 1024    3]
LEARNING_MOMENTUM              0.9
LEARNING_RATE                  0.002
MASK_POOL_SIZE                 14
MASK_SHAPE                     [28, 28]
MAX_GT_INSTANCES               100
MEAN_PIXEL                     [ 123.7  116.8  103.9]
MINI_MASK_SHAPE                (56, 56)
NAME                           coco
NUM_CLASSES                    81
POOL_SIZE                      7
POST_NMS_ROIS_INFERENCE        1000
POST_NMS_ROIS_TRAINING         2000
ROI_POSITIVE_RATIO             0.33
RPN_ANCHOR_RATIOS              [0.5, 1, 2]
RPN_ANCHOR_SCALES              (32, 64, 128, 256, 512)
RPN_ANCHOR_STRIDE              2
RPN_BBOX_STD_DEV               [ 0.1  0.1  0.2  0.2]
RPN_TRAIN_ANCHORS_PER_IMAGE    256
STEPS_PER_EPOCH                1000
TRAIN_ROIS_PER_IMAGE           128
USE_MINI_MASK                  True
USE_RPN_ROIS                   True
VALIDATION_STEPS               50
WEIGHT_DECAY                   0.0001

生成模型並載入訓練權重

# Create model object in inference mode.
model = modellib.MaskRCNN(mode="inference", model_dir=MODEL_DIR, config=config)

# Load weights trained on MS-COCO
model.load_weights(COCO_MODEL_PATH, by_name=True)

種類名稱

# COCO Class names
# Index of the class in the list is its ID. For example, to get ID of
# the teddy bear class, use: class_names.index('teddy bear')
class_names = ['BG', 'person', 'bicycle', 'car', 'motorcycle', 'airplane',
               'bus', 'train', 'truck', 'boat', 'traffic light',
               'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird',
               'cat', 'dog', 'horse', 'sheep', 'cow', 'elephant', 'bear',
               'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie',
               'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball',
               'kite', 'baseball bat', 'baseball glove', 'skateboard',
               'surfboard', 'tennis racket', 'bottle', 'wine glass', 'cup',
               'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple',
               'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza',
               'donut', 'cake', 'chair', 'couch', 'potted plant', 'bed',
               'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
               'keyboard', 'cell phone', 'microwave', 'oven', 'toaster',
               'sink', 'refrigerator', 'book', 'clock', 'vase', 'scissors',
               'teddy bear', 'hair drier', 'toothbrush']

隨機選取圖片測試

# Load a random image from the images folder
file_names = next(os.walk(IMAGE_DIR))[2]
image = skimage.io.imread(os.path.join(IMAGE_DIR, random.choice(file_names)))

# Run detection
results = model.detect([image], verbose=1)

# Visualize results
r = results[0]
visualize.display_instances(image, r['rois'], r['masks'], r['class_ids'], 
                            class_names, r['scores'])

Processing 1 images
image                    shape: (476, 640, 3)         min:    0.00000  max:  255.00000
molded_images            shape: (1, 1024, 1024, 3)    min: -123.70000  max:  120.30000
image_metas              shape: (1, 89)               min:    0.00000  max: 1024.00000

上面的demo.ipynb是對mask rcnn專案的簡單運用，當你成功實現之後便是對這個專案的簡單入門。

2. inspect_model.ipynb

import os
import sys
import random
import math
import re
import time
import numpy as np
import tensorflow as tf
import matplotlib
import matplotlib.pyplot as plt
import matplotlib.patches as patches

# Root directory of the project
ROOT_DIR = os.path.abspath("../../")  #專案資料夾

# Import Mask RCNN
sys.path.append(ROOT_DIR)  # To find local version of the library
from mrcnn import utils
from mrcnn import visualize
from mrcnn.visualize import display_images
import mrcnn.model as modellib
from mrcnn.model import log

%matplotlib inline 

# Directory to save logs and trained model
MODEL_DIR = os.path.join(ROOT_DIR, "logs")

# Local path to trained weights file
COCO_MODEL_PATH = os.path.join(ROOT_DIR, "mask_rcnn_coco.h5")
# Download COCO trained weights from Releases if needed
if not os.path.exists(COCO_MODEL_PATH):
    utils.download_trained_weights(COCO_MODEL_PATH)

# Path to Shapes trained weights
#SHAPES_MODEL_PATH = os.path.join(ROOT_DIR, "mask_rcnn_shapes.h5")

Configurations

# MS COCO Dataset
import coco
config = coco.CocoConfig()
COCO_DIR = "path to COCO dataset"  # TODO: enter value here

# Override the training configurations with a few
# changes for inferencing.
class InferenceConfig(config.__class__):
    # Run detection on one image at a time
    GPU_COUNT = 1
    IMAGES_PER_GPU = 1
config = InferenceConfig()
config.display()

Notebook Preferences

# Device to load the neural network on.
# Useful if you're training a model on the same 
# machine, in which case use CPU and leave the
# GPU for training.
DEVICE = "/cpu:0"  # /cpu:0 or /gpu:0

# Inspect the model in training or inference modes
# values: 'inference' or 'training'
# TODO: code for 'training' test mode not ready yet
TEST_MODE = "inference"def get_ax(rows=1, cols=1, size=16):
    """Return a Matplotlib Axes array to be used in
    all visualizations in the notebook. Provide a
    central point to control graph sizes.
    
    Adjust the size attribute to control how big to render images
    """
    _, ax = plt.subplots(rows, cols, figsize=(size*cols, size*rows))
    return ax

Load Validation Dataset

# Build validation dataset
if config.NAME == 'shapes':
    dataset = shapes.ShapesDataset()
    dataset.load_shapes(500, config.IMAGE_SHAPE[0], config.IMAGE_SHAPE[1])
elif config.NAME == "coco":
    dataset = coco.CocoDataset()
    dataset.load_coco(COCO_DIR, "minival") #將下載好的 train2014 val2014 test2014 圖片解壓到coco資料夾

# Must call before using the dataset
dataset.prepare()

print("Images: {}\nClasses: {}".format(len(dataset.image_ids), dataset.class_names

loading annotations into memory...
Done (t=4.86s)
creating index...
index created!
Images: 35185
Classes: ['BG', 'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train', 'truck', 'boat', 'traffic light', 'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow', 'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie', 'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball', 'kite', 'baseball bat', 'baseball glove', 'skateboard', 'surfboard', 'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple', 'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', 'potted plant', 'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote', 'keyboard', 'cell phone', 'microwave', 'oven', 'toaster', 'sink', 'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear', 'hair drier', 'toothbrush']

如果報錯顯示 ../coco/annotations缺少instances_valminusminival2014.json和instances_minival2014.json，就在終端輸入以下命令：

wget http://images.cocodataset.org/annotations/annotations_trainval2014.zip #解壓放到對應資料夾即可

Load Model

# Create model in inference mode
with tf.device(DEVICE):
    model = modellib.MaskRCNN(mode="inference", model_dir=MODEL_DIR,
                              config=config)

# Set weights file path
if config.NAME == "shapes":
    weights_path = SHAPES_MODEL_PATH
elif config.NAME == "coco":
    weights_path = COCO_MODEL_PATH
# Or, uncomment to load the last model you trained
# weights_path = model.find_last()[1]

# Load weights
print("Loading weights ", weights_path)
model.load_weights(weights_path, by_name=True)

Run Detection¶

image_id = random.choice(dataset.image_ids)
image, image_meta, gt_class_id, gt_bbox, gt_mask =\
    modellib.load_image_gt(dataset, config, image_id, use_mini_mask=False)
info = dataset.image_info[image_id]
print("image ID: {}.{} ({}) {}".format(info["source"], info["id"], image_id, 
                                       dataset.image_reference(image_id)))
# Run object detection
results = model.detect([image], verbose=1)

# Display results
ax = get_ax(1)
r = results[0]
visualize.display_instances(image, r['rois'], r['masks'], r['class_ids'], 
                            dataset.class_names, r['scores'], ax=ax,
                            title="Predictions")
log("gt_class_id", gt_class_id)
log("gt_bbox", gt_bbox)
log("gt_mask", gt_mask)

image ID: coco.392144 (34940) http://cocodataset.org/#explore?id=392144
Processing 1 images
image                    shape: (1024, 1024, 3)       min:    0.00000  max:  255.00000
molded_images            shape: (1, 1024, 1024, 3)    min: -123.70000  max:  151.10000
image_metas              shape: (1, 89)               min:    0.00000  max: 1024.00000
gt_class_id              shape: (10,)                 min:    1.00000  max:   40.00000
gt_bbox                  shape: (10, 5)               min:    0.00000  max: 1024.00000
gt_mask                  shape: (1024, 1024, 10)      min:    0.00000  max:    1.00000

Precision-Recall

# Draw precision-recall curve
AP, precisions, recalls, overlaps = utils.compute_ap(gt_bbox, gt_class_id, gt_mask,
                                          r['rois'], r['class_ids'], r['scores'], r['masks'])
visualize.plot_precision_recall(AP, precisions, recalls)

# Grid of ground truth objects and their predictions
visualize.plot_overlaps(gt_class_id, r['class_ids'], r['scores'],
                        overlaps, dataset.class_names)

Compute mAP @ IoU=50 on Batch of Images

# Compute VOC-style Average Precision
def compute_batch_ap(image_ids):
    APs = []
    for image_id in image_ids:
        # Load image
        image, image_meta, gt_class_id, gt_bbox, gt_mask =\
            modellib.load_image_gt(dataset, config,
                                   image_id, use_mini_mask=False)
        # Run object detection
        results = model.detect([image], verbose=0)
        # Compute AP
        r = results[0]
        AP, precisions, recalls, overlaps =\
            utils.compute_ap(gt_bbox, gt_class_id, gt_mask,
                              r['rois'], r['class_ids'], r['scores'], r['masks'])
        APs.append(AP)
    return APs

# Pick a set of random images
image_ids = np.random.choice(dataset.image_ids, 10)
APs = compute_batch_ap(image_ids)
print("mAP @ IoU=50: ", np.mean(APs))

mAP @ IoU=50: 0.656323084916

Step by Step Prediction

1.a RPN Targets

# Generate RPN trainig targets
# target_rpn_match is 1 for positive anchors, -1 for negative anchors
# and 0 for neutral anchors.
target_rpn_match, target_rpn_bbox = modellib.build_rpn_targets(
    image.shape, model.anchors, gt_class_id, gt_bbox, model.config)
log("target_rpn_match", target_rpn_match)
log("target_rpn_bbox", target_rpn_bbox)

positive_anchor_ix = np.where(target_rpn_match[:] == 1)[0]
negative_anchor_ix = np.where(target_rpn_match[:] == -1)[0]
neutral_anchor_ix = np.where(target_rpn_match[:] == 0)[0]
positive_anchors = model.anchors[positive_anchor_ix]
negative_anchors = model.anchors[negative_anchor_ix]
neutral_anchors = model.anchors[neutral_anchor_ix]
log("positive_anchors", positive_anchors)
log("negative_anchors", negative_anchors)
log("neutral anchors", neutral_anchors)

# Apply refinement deltas to positive anchors
refined_anchors = utils.apply_box_deltas(
    positive_anchors,
    target_rpn_bbox[:positive_anchors.shape[0]] * model.config.RPN_BBOX_STD_DEV)
log("refined_anchors", refined_anchors, )

target_rpn_match         shape: (65472,)              min:   -1.00000  max:    1.00000
target_rpn_bbox          shape: (256, 4)              min:   -5.19860  max:    2.59641
positive_anchors         shape: (14, 4)               min:    5.49033  max:  973.25483
negative_anchors         shape: (242, 4)              min:  -22.62742  max: 1038.62742
neutral anchors          shape: (65216, 4)            min: -362.03867  max: 1258.03867
refined_anchors          shape: (14, 4)               min:    0.00000  max: 1023.99994

# Display positive anchors before refinement (dotted) and
# after refinement (solid).
visualize.draw_boxes(image, boxes=positive_anchors, refined_boxes=refined_anchors, ax=get_ax())

1.b RPN Prediction

# Run RPN sub-graph
pillar = model.keras_model.get_layer("ROI").output  # node to start searching from

# TF 1.4 introduces a new version of NMS. Search for both names to support TF 1.3 and 1.4
nms_node = model.ancestor(pillar, "ROI/rpn_non_max_suppression:0")
if nms_node is None:
    nms_node = model.ancestor(pillar, "ROI/rpn_non_max_suppression/NonMaxSuppressionV2:0")

rpn = model.run_graph([image], [
    ("rpn_class", model.keras_model.get_layer("rpn_class").output),
    ("pre_nms_anchors", model.ancestor(pillar, "ROI/pre_nms_anchors:0")),
    ("refined_anchors", model.ancestor(pillar, "ROI/refined_anchors:0")),
    ("refined_anchors_clipped", model.ancestor(pillar, "ROI/refined_anchors_clipped:0")),
    ("post_nms_anchor_ix", nms_node),
    ("proposals", model.keras_model.get_layer("ROI").output),
])

rpn_class                shape: (1, 65472, 2)         min:    0.00000  max:    1.00000
pre_nms_anchors          shape: (1, 10000, 4)         min: -362.03867  max: 1258.03870
refined_anchors          shape: (1, 10000, 4)         min: -1385.67920  max: 2212.44043
refined_anchors_clipped  shape: (1, 10000, 4)         min:    0.00000  max: 1024.00000
post_nms_anchor_ix       shape: (1000,)               min:    0.00000  max: 1477.00000
proposals                shape: (1, 1000, 4)          min:    0.00000  max:    1.00000

# Show top anchors by score (before refinement)
limit = 100
sorted_anchor_ids = np.argsort(rpn['rpn_class'][:,:,1].flatten())[::-1]
visualize.draw_boxes(image, boxes=model.anchors[sorted_anchor_ids[:limit]], ax=get_ax())

# Show top anchors with refinement. Then with clipping to image boundaries
limit = 50
ax = get_ax(1, 2)
visualize.draw_boxes(image, boxes=rpn["pre_nms_anchors"][0, :limit], 
           refined_boxes=rpn["refined_anchors"][0, :limit], ax=ax[0])
visualize.draw_boxes(image, refined_boxes=rpn["refined_anchors_clipped"][0, :limit], ax=ax[1])

# Show refined anchors after non-max suppression
limit = 50
ixs = rpn["post_nms_anchor_ix"][:limit]
visualize.draw_boxes(image, refined_boxes=rpn["refined_anchors_clipped"][0, ixs], ax=get_ax())

# Show final proposals
# These are the same as the previous step (refined anchors 
# after NMS) but with coordinates normalized to [0, 1] range.
limit = 50
# Convert back to image coordinates for display
h, w = config.IMAGE_SHAPE[:2]
proposals = rpn['proposals'][0, :limit] * np.array([h, w, h, w])
visualize.draw_boxes(image, refined_boxes=proposals, ax=get_ax())

Stage 2: Proposal Classification

2.a Proposal Classification

# Get input and output to classifier and mask heads.
mrcnn = model.run_graph([image], [
    ("proposals", model.keras_model.get_layer("ROI").output),
    ("probs", model.keras_model.get_layer("mrcnn_class").output),
    ("deltas", model.keras_model.get_layer("mrcnn_bbox").output),
    ("masks", model.keras_model.get_layer("mrcnn_mask").output),
    ("detections", model.keras_model.get_layer("mrcnn_detection").output),
])

# Get detection class IDs. Trim zero padding.
det_class_ids = mrcnn['detections'][0, :, 4].astype(np.int32)
det_count = np.where(det_class_ids == 0)[0][0]
det_class_ids = det_class_ids[:det_count]
detections = mrcnn['detections'][0, :det_count]

print("{} detections: {}".format(
    det_count, np.array(dataset.class_names)[det_class_ids]))

captions = ["{} {:.3f}".format(dataset.class_names[int(c)], s) if c > 0 else ""
            for c, s in zip(detections[:, 4], detections[:, 5])]
visualize.draw_boxes(
    image, 
    refined_boxes=utils.denorm_boxes(detections[:, :4], image.shape[:2]),
    visibilities=[2] * len(detections),
    captions=captions, title="Detections",
    ax=get_ax())

8 detections: ['person' 'person' 'person' 'person' 'person' 'airplane' 'airplane' 'car']

# Proposals are in normalized coordinates. Scale them
# to image coordinates.
h, w = config.IMAGE_SHAPE[:2]
proposals = np.around(mrcnn["proposals"][0] * np.array([h, w, h, w])).astype(np.int32)

# Class ID, score, and mask per proposal
roi_class_ids = np.argmax(mrcnn["probs"][0], axis=1)
roi_scores = mrcnn["probs"][0, np.arange(roi_class_ids.shape[0]), roi_class_ids]
roi_class_names = np.array(dataset.class_names)[roi_class_ids]
roi_positive_ixs = np.where(roi_class_ids > 0)[0]

# How many ROIs vs empty rows?
print("{} Valid proposals out of {}".format(np.sum(np.any(proposals, axis=1)), proposals.shape[0]))
print("{} Positive ROIs".format(len(roi_positive_ixs)))

# Class counts
print(list(zip(*np.unique(roi_class_names, return_counts=True))))

1000 Valid proposals out of 1000
71 Positive ROIs
[('BG', 929), ('airplane', 23), ('car', 11), ('person', 37)]

# Show final detections
ixs = np.arange(len(keep))  # Display all
# ixs = np.random.randint(0, len(keep), 10)  # Display random sample
captions = ["{} {:.3f}".format(dataset.class_names[c], s) if c > 0 else ""
            for c, s in zip(roi_class_ids[keep][ixs], roi_scores[keep][ixs])]
visualize.draw_boxes(
    image, boxes=proposals[keep][ixs],
    refined_boxes=refined_proposals[keep][ixs],
    visibilities=np.where(roi_class_ids[keep][ixs] > 0, 1, 0),
    captions=captions, title="Detections after NMS",
    ax=get_ax())

Stage 3: Generating Masks

3.a Mask Targets

display_images(np.transpose(gt_mask, [2, 0, 1]), cmap="Blues")

display_images(det_mask_specific[:4] * 255, cmap="Blues", interpolation="none")

display_images(det_masks[:4] * 255, cmap="Blues", interpolation="none")

Visualize Activations

# Get activations of a few sample layers
activations = model.run_graph([image], [
    ("input_image",        model.keras_model.get_layer("input_image").output),
    ("res4w_out",          model.keras_model.get_layer("res4w_out").output),  # for resnet100
    ("rpn_bbox",           model.keras_model.get_layer("rpn_bbox").output),
    ("roi",                model.keras_model.get_layer("ROI").output),
])

# Input image (normalized)
_ = plt.imshow(modellib.unmold_image(activations["input_image"][0],config))

# Backbone feature map
display_images(np.transpose(activations["res4w_out"][0,:,:,:4], [2, 0, 1]))

# Histograms of RPN bounding box deltas
plt.figure(figsize=(12, 3))
plt.subplot(1, 4, 1)
plt.title("dy")
_ = plt.hist(activations["rpn_bbox"][0,:,0], 50)
plt.subplot(1, 4, 2)
plt.title("dx")
_ = plt.hist(activations["rpn_bbox"][0,:,1], 50)
plt.subplot(1, 4, 3)
plt.title("dw")
_ = plt.hist(activations["rpn_bbox"][0,:,2], 50)
plt.subplot(1, 4, 4)
plt.title("dh")
_ = plt.hist(activations["rpn_bbox"][0,:,3], 50)

# Histograms of RPN bounding box deltas
plt.figure(figsize=(12, 3))
plt.subplot(1, 4, 1)
plt.title("dy")
_ = plt.hist(activations["rpn_bbox"][0,:,0], 50)
plt.subplot(1, 4, 2)
plt.title("dx")
_ = plt.hist(activations["rpn_bbox"][0,:,1], 50)
plt.subplot(1, 4, 3)
plt.title("dw")
_ = plt.hist(activations["rpn_bbox"][0,:,2], 50)
plt.subplot(1, 4, 4)
plt.title("dh")
_ = plt.hist(activations["rpn_bbox"][0,:,3], 50)

# Distribution of y, x coordinates of generated proposals
plt.figure(figsize=(10, 5))
plt.subplot(1, 2, 1)
plt.title("y1, x1")
plt.scatter(activations["roi"][0,:,0], activations["roi"][0,:,1])
plt.subplot(1, 2, 2)
plt.title("y2, x2")
plt.scatter(activations["roi"][0,:,2], activations["roi"][0,:,3])
plt.show()