faster rcnn源碼解讀（六）之minibatch

轉載自：faster rcnn源碼解讀（六）之minibatch - 野孩子的專欄 - 博客頻道 - CSDN.NET
http://blog.csdn.net/u010668907/article/details/51945917

faster rcnn用python版本的https://github.com/rbgirshick/py-faster-rcnn

minibatch源碼：https://github.com/rbgirshick/py-faster-rcnn/blob/master/lib/roi_data_layer/minibatch.py

源碼：

# --------------------------------------------------------
# Fast R-CNN
# Copyright (c) 2015 Microsoft
# Licensed under The MIT License [see LICENSE for details]
# Written by Ross Girshick
# --------------------------------------------------------

"""Compute minibatch blobs for training a Fast R-CNN network."""

import numpy as np
import numpy.random as npr
import cv2
from fast_rcnn.config import cfg
from utils.blob import prep_im_for_blob, im_list_to_blob

def get_minibatch(roidb, num_classes):
    """Given a roidb, construct a minibatch sampled from it."""
    num_images = len(roidb)
    # Sample random scales to use for each image in this batch
    random_scale_inds = npr.randint(0, high=len(cfg.TRAIN.SCALES),
                                    size=num_images)#隨機索引組成的numpy，大小是roidb的長度
    assert(cfg.TRAIN.BATCH_SIZE % num_images == 0), \
        'num_images ({}) must divide BATCH_SIZE ({})'. \
        format(num_images, cfg.TRAIN.BATCH_SIZE)
    rois_per_image = cfg.TRAIN.BATCH_SIZE / num_images
    fg_rois_per_image = np.round(cfg.TRAIN.FG_FRACTION * rois_per_image)

    # Get the input image blob, formatted for caffe
    im_blob, im_scales = _get_image_blob(roidb, random_scale_inds)

    blobs = {'data': im_blob}

    if cfg.TRAIN.HAS_RPN:
        assert len(im_scales) == 1, "Single batch only"
        assert len(roidb) == 1, "Single batch only"
        # gt boxes: (x1, y1, x2, y2, cls)
        gt_inds = np.where(roidb[0]['gt_classes'] != 0)[0]
        gt_boxes = np.empty((len(gt_inds), 5), dtype=np.float32)
        gt_boxes[:, 0:4] = roidb[0]['boxes'][gt_inds, :] * im_scales[0]
        gt_boxes[:, 4] = roidb[0]['gt_classes'][gt_inds]
        blobs['gt_boxes'] = gt_boxes
        blobs['im_info'] = np.array(
            [[im_blob.shape[2], im_blob.shape[3], im_scales[0]]],
            dtype=np.float32)
    else: # not using RPN
        # Now, build the region of interest and label blobs
        rois_blob = np.zeros((0, 5), dtype=np.float32)
        labels_blob = np.zeros((0), dtype=np.float32)
        bbox_targets_blob = np.zeros((0, 4 * num_classes), dtype=np.float32)
        bbox_inside_blob = np.zeros(bbox_targets_blob.shape, dtype=np.float32)
        # all_overlaps = []
        for im_i in xrange(num_images):
            labels, overlaps, im_rois, bbox_targets, bbox_inside_weights \
                = _sample_rois(roidb[im_i], fg_rois_per_image, rois_per_image,
                               num_classes)

            # Add to RoIs blob
            rois = _project_im_rois(im_rois, im_scales[im_i])
            batch_ind = im_i * np.ones((rois.shape[0], 1))
            rois_blob_this_image = np.hstack((batch_ind, rois))
            rois_blob = np.vstack((rois_blob, rois_blob_this_image))

            # Add to labels, bbox targets, and bbox loss blobs
            labels_blob = np.hstack((labels_blob, labels))
            bbox_targets_blob = np.vstack((bbox_targets_blob, bbox_targets))
            bbox_inside_blob = np.vstack((bbox_inside_blob, bbox_inside_weights))
            # all_overlaps = np.hstack((all_overlaps, overlaps))

        # For debug visualizations
        # _vis_minibatch(im_blob, rois_blob, labels_blob, all_overlaps)

        blobs['rois'] = rois_blob
        blobs['labels'] = labels_blob

        if cfg.TRAIN.BBOX_REG:
            blobs['bbox_targets'] = bbox_targets_blob
            blobs['bbox_inside_weights'] = bbox_inside_blob
            blobs['bbox_outside_weights'] = \
                np.array(bbox_inside_blob > 0).astype(np.float32)

    return blobs

def _sample_rois(roidb, fg_rois_per_image, rois_per_image, num_classes):
    """Generate a random sample of RoIs comprising foreground and background
    examples.
    """
    # label = class RoI has max overlap with
    labels = roidb['max_classes']
    overlaps = roidb['max_overlaps']
    rois = roidb['boxes']

    # Select foreground RoIs as those with >= FG_THRESH overlap
    fg_inds = np.where(overlaps >= cfg.TRAIN.FG_THRESH)[0]
    # Guard against the case when an image has fewer than fg_rois_per_image
    # foreground RoIs
    fg_rois_per_this_image = np.minimum(fg_rois_per_image, fg_inds.size)
    # Sample foreground regions without replacement
    if fg_inds.size > 0:
        fg_inds = npr.choice(
                fg_inds, size=fg_rois_per_this_image, replace=False)

    # Select background RoIs as those within [BG_THRESH_LO, BG_THRESH_HI)
    bg_inds = np.where((overlaps < cfg.TRAIN.BG_THRESH_HI) &
                       (overlaps >= cfg.TRAIN.BG_THRESH_LO))[0]
    # Compute number of background RoIs to take from this image (guarding
    # against there being fewer than desired)
    bg_rois_per_this_image = rois_per_image - fg_rois_per_this_image
    bg_rois_per_this_image = np.minimum(bg_rois_per_this_image,
                                        bg_inds.size)
    # Sample foreground regions without replacement
    if bg_inds.size > 0:
        bg_inds = npr.choice(
                bg_inds, size=bg_rois_per_this_image, replace=False)

    # The indices that we're selecting (both fg and bg)
    keep_inds = np.append(fg_inds, bg_inds)
    # Select sampled values from various arrays:
    labels = labels[keep_inds]
    # Clamp labels for the background RoIs to 0
    labels[fg_rois_per_this_image:] = 0
    overlaps = overlaps[keep_inds]
    rois = rois[keep_inds]

    bbox_targets, bbox_inside_weights = _get_bbox_regression_labels(
            roidb['bbox_targets'][keep_inds, :], num_classes)

    return labels, overlaps, rois, bbox_targets, bbox_inside_weights

def _get_image_blob(roidb, scale_inds):
    """Builds an input blob from the images in the roidb at the specified
    scales.
    """
    num_images = len(roidb)
    processed_ims = []
    im_scales = []
    for i in xrange(num_images):
        im = cv2.imread(roidb[i]['image'])
        if roidb[i]['flipped']:
            im = im[:, ::-1, :]
        target_size = cfg.TRAIN.SCALES[scale_inds[i]]
        im, im_scale = prep_im_for_blob(im, cfg.PIXEL_MEANS, target_size,
                                        cfg.TRAIN.MAX_SIZE)prep_im_for_blob： util的blob.py中；用於將圖片平均後縮放。#im_scales： 每張圖片的縮放率

        #  cfg.PIXEL_MEANS： 原始圖片會集體減去該值達到mean
        im_scales.append(im_scale)
        processed_ims.append(im)

    # Create a blob to hold the input images
    blob = im_list_to_blob(processed_ims)#將以list形式存放的圖片數據處理成(batch elem, channel, height, width)的im_blob形式，height，width用的是此次計算所有圖片的最大值

    return blob, im_scales#blob是一個字典，與name_to_top對應，方便把blob數據放進top

def _project_im_rois(im_rois, im_scale_factor):
    """Project image RoIs into the rescaled training image."""
    rois = im_rois * im_scale_factor
    return rois

def _get_bbox_regression_labels(bbox_target_data, num_classes):
    """Bounding-box regression targets are stored in a compact form in the
    roidb.

    This function expands those targets into the 4-of-4*K representation used
    by the network (i.e. only one class has non-zero targets). The loss weights
    are similarly expanded.

    Returns:
        bbox_target_data (ndarray): N x 4K blob of regression targets
        bbox_inside_weights (ndarray): N x 4K blob of loss weights
    """
    clss = bbox_target_data[:, 0]
    bbox_targets = np.zeros((clss.size, 4 * num_classes), dtype=np.float32)
    bbox_inside_weights = np.zeros(bbox_targets.shape, dtype=np.float32)
    inds = np.where(clss > 0)[0]
    for ind in inds:
        cls = clss[ind]
        start = 4 * cls
        end = start + 4
        bbox_targets[ind, start:end] = bbox_target_data[ind, 1:]
        bbox_inside_weights[ind, start:end] = cfg.TRAIN.BBOX_INSIDE_WEIGHTS
    return bbox_targets, bbox_inside_weights

def _vis_minibatch(im_blob, rois_blob, labels_blob, overlaps):
    """Visualize a mini-batch for debugging."""
    import matplotlib.pyplot as plt
    for i in xrange(rois_blob.shape[0]):
        rois = rois_blob[i, :]
        im_ind = rois[0]
        roi = rois[1:]
        im = im_blob[im_ind, :, :, :].transpose((1, 2, 0)).copy()
        im += cfg.PIXEL_MEANS
        im = im[:, :, (2, 1, 0)]
        im = im.astype(np.uint8)
        cls = labels_blob[i]
        plt.imshow(im)
        print 'class: ', cls, ' overlap: ', overlaps[i]
        plt.gca().add_patch(
            plt.Rectangle((roi[0], roi[1]), roi[2] - roi[0],
                          roi[3] - roi[1], fill=False,
                          edgecolor='r', linewidth=3)
            )
        plt.show()

solver.step(1)-》reshape-》forward-》_get_next_minbatch-》_get_next_minbatch_inds-》(前面在layers裏,現在進入minibatch組建真正的blob)get_minibatch

4.1 cfg.TRAIN.SCALES： 圖片被縮放的target_size列表

  4.2 random_scale_inds：列表的隨機索引組成的numpy，大小是roidb的長度

  4.3 cfg.PIXEL_MEANS： 原始圖片會集體減去該值達到mean

  4.4 im_scales： 每張圖片的縮放率

    縮放率的求法： im_scales = target_size/min(width, height);

              if im_scales*max(width, height) > cfg.TRAIN.MAX_SIZE  

                    im_scales = cfg.TRAIN.MAX_SIZE * max(width, height)

  4.5 prep_im_for_blob： util的blob.py中；用於將圖片平均後縮放。

 4.6 im_list_to_blob(ims): 將以list形式存放的圖片數據處理成(batch elem, channel, height, width)的im_blob形式，height，width用的是此次計算所有圖片的最大值

4.7 blob是一個字典：data，一個batch的處理過的所有圖片數據，即上面的im_blob；

                       im_info， [im_blob.shape[2], im_blob.shape[3], im_scales[0]]

                       gt_boxes， 是前四列是box的值，第五列是box的類別。

                              box=原box*im_scales

  blob與name_to_top對應，方便把blob數據放進top

！！！minibatch.py中34行的代碼表明，batchsize即cfg.TRAIN.IMS_PER_BATCH只能是1