徐静
paper: Mask R-CNN
Mask R-CNN的Python3, Keras和TensorFlow实现,该网络基于Feature Pyramid Network(FPN)和ResNet101 backbone
该Repository实现了:
- 基于FPN和ResNet101的Mask R-CNN网络.
- 训练基于COCO标注数据样式的标注
- 预训练模型
- Jupyter notebooks 可是化整个训练和推断过程
- 多GPU并行训练
- 评价指标AP
- 训练自己数据及的样例
相关文件说明:
- demo.ipynb 用预训练的MS COCO数据集分割自己数据的demo
- train_shapes.ipynb 如何使用MS R-CNN训练自己的数据集,以toy data Shapes为例
- (model.py, utils.py, config.py): Mask R-CNN的实现
- inspect_data.ipynb. 数据预处理每一步的可视化
- inspect_model.ipynb 模型训练过程的可视化
- inspect_weights.ipynb 训练好的模型的可视化
使用labelme做标注,标注后转化为COCO样式的数据,该过程请参考labelme教程,在此不赘述,标注的训练数据为专业的医学消化内镜数据,因涉及数据安全和隐私的保护,不会提供和公开该标注数据集。
详细参考./samples/water/coco.py
- 重写config文件
class CocoConfig(Config):
"""Configuration for training on MS COCO.
Derives from the base Config class and overrides values specific
to the COCO dataset.
"""
# Give the configuration a recognizable name
NAME = "coco"
# We use a GPU with 12GB memory, which can fit two images.
# Adjust down if you use a smaller GPU.
# IMAGES_PER_GPU = 8
IMAGES_PER_GPU = 2
# Uncomment to train on 8 GPUs (default is 1)
GPU_COUNT = 1
# Number of classes (including background)
NUM_CLASSES = 1 + 2 # COCO has 80 classes
# Backbone network architecture
# Supported values are: resnet50, resnet101
BACKBONE = "resnet50"
# Input image resizing
# Random crops of size 512x512
IMAGE_RESIZE_MODE = "crop"
IMAGE_MIN_DIM = 512
IMAGE_MAX_DIM = 512
IMAGE_MIN_SCALE = 2.0
# Length of square anchor side in pixels
RPN_ANCHOR_SCALES = (8, 16, 32, 64, 128)
关于config的介绍请参考./mrcnn/config.py
- 重写dataset
class CocoDataset(utils.Dataset):
def load_coco(self, dataset_dir, subset,class_ids=None,class_map=None, return_coco=False):
"""Load a subset of the COCO dataset.
dataset_dir: The root directory of the COCO dataset.
subset: What to load (train, val)
class_ids: If provided, only loads images that have the given classes.
class_map: TODO: Not implemented yet. Supports maping classes from
different datasets to the same class ID.
return_coco: If True, returns the COCO object.
"""
# 将自己的标注和图像文件的路径修改成自己的
# coco = COCO("{}/annotations/instances_{}{}.json".format(dataset_dir, subset, year))
coco = COCO("{}/annotations/annotations_{}.json".format(dataset_dir, subset))
# image_dir = "{}/{}".format(dataset_dir, subset)
image_dir = "{}/images".format(dataset_dir)
# Load all classes or a subset?
# 注意将标注文件中__background__类别去掉
if not class_ids:
# All classes
class_ids = sorted(coco.getCatIds())
# All images or a subset?
if class_ids:
image_ids = []
for id_ in class_ids:
image_ids.extend(list(coco.getImgIds(catIds=[id_])))
# Remove duplicates
image_ids = list(set(image_ids))
else:
# All images
image_ids = list(coco.imgs.keys())
# Add classes
for i in class_ids:
self.add_class("coco", i, coco.loadCats(i)[0]["name"])
# Add images
for i in image_ids:
self.add_image(
"coco", image_id=i,
path=os.path.join(image_dir, coco.imgs[i]['file_name']),
width=coco.imgs[i]["width"],
height=coco.imgs[i]["height"],
annotations=coco.loadAnns(coco.getAnnIds(
imgIds=[i], catIds=class_ids, iscrowd=None)))
if return_coco:
return coco
def load_mask(self, image_id):
"""Load instance masks for the given image.
Different datasets use different ways to store masks. This
function converts the different mask format to one format
in the form of a bitmap [height, width, instances].
Returns:
masks: A bool array of shape [height, width, instance count] with
one mask per instance.
class_ids: a 1D array of class IDs of the instance masks.
"""
# If not a COCO image, delegate to parent class.
image_info = self.image_info[image_id]
if image_info["source"] != "coco":
return super(CocoDataset, self).load_mask(image_id)
instance_masks = []
class_ids = []
annotations = self.image_info[image_id]["annotations"]
# Build mask of shape [height, width, instance_count] and list
# of class IDs that correspond to each channel of the mask.
for annotation in annotations:
class_id = self.map_source_class_id(
"coco.{}".format(annotation['category_id']))
if class_id:
m = self.annToMask(annotation, image_info["height"],
image_info["width"])
# Some objects are so small that they're less than 1 pixel area
# and end up rounded out. Skip those objects.
if m.max() < 1:
continue
# Is it a crowd? If so, use a negative class ID.
if annotation['iscrowd']:
# Use negative class ID for crowds
class_id *= -1
# For crowd masks, annToMask() sometimes returns a mask
# smaller than the given dimensions. If so, resize it.
if m.shape[0] != image_info["height"] or m.shape[1] != image_info["width"]:
m = np.ones([image_info["height"], image_info["width"]], dtype=bool)
instance_masks.append(m)
class_ids.append(class_id)
# Pack instance masks into an array
if class_ids:
mask = np.stack(instance_masks, axis=2).astype(np.bool)
class_ids = np.array(class_ids, dtype=np.int32)
return mask, class_ids
else:
# Call super class to return an empty mask
return super(CocoDataset, self).load_mask(image_id)
# def image_reference(self, image_id):
# """Return a link to the image in the COCO Website."""
# info = self.image_info[image_id]
# if info["source"] == "coco":
# return "http://cocodataset.org/#explore?id={}".format(info["id"])
# else:
# super(CocoDataset, self).image_reference(image_id)
# The following two functions are from pycocotools with a few changes.
def annToRLE(self, ann, height, width):
"""
Convert annotation which can be polygons, uncompressed RLE to RLE.
:return: binary mask (numpy 2D array)
"""
segm = ann['segmentation']
if isinstance(segm, list):
# polygon -- a single object might consist of multiple parts
# we merge all parts into one mask rle code
rles = maskUtils.frPyObjects(segm, height, width)
rle = maskUtils.merge(rles)
elif isinstance(segm['counts'], list):
# uncompressed RLE
rle = maskUtils.frPyObjects(segm, height, width)
else:
# rle
rle = ann['segmentation']
return rle
def annToMask(self, ann, height, width):
"""
Convert annotation which can be polygons, uncompressed RLE, or RLE to binary mask.
:return: binary mask (numpy 2D array)
"""
rle = self.annToRLE(ann, height, width)
m = maskUtils.decode(rle)
return m
请参考./samples/water/detection.py
,因原Mask R-CNN使用skimage和matplotlib处理数据,我们实现了通过OpenCV处理数据,并将最终的识别结果通过OpenCV保存和抛出。
请参考./samples/water/detection_video.py
,原Mask R-CNN中并未提供视频推断的源码,我们基于OpenCV实现了训练Mask R-CNN模型的视频推断过程,并计算推断的帧率(FPS)
训练模型
# 基于pre-trained COCO weights训练新的模型
python3 coco.py train --dataset=./myData/coco/ --model=coco
#基于预训练的 ImageNet weights,训练新的模型
python3 coco.py train --dataset=./myData/coco/ --model=imagenet
# 断点训练1
python3 coco.py train --dataset=/path/to/coco/ --model=/path/to/weights.h5
# 断点训练2
python3 coco.py train --dataset=/path/to/coco/ --model=last
模型推断
# 图像
python3 detection.py
# 视频
python3 detection_video.py
@misc{matterport_maskrcnn_2017,
title={Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow},
author={Waleed Abdulla},
year={2017},
publisher={Github},
journal={GitHub repository},
howpublished={\url{https://github.com/matterport/Mask_RCNN}},
}