infer.py 9.2 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
# Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import os
import glob

import numpy as np
from PIL import Image

W
wangguanzhong 已提交
25

26 27 28 29 30
def set_paddle_flags(**kwargs):
    for key, value in kwargs.items():
        if os.environ.get(key, None) is None:
            os.environ[key] = str(value)

W
wangguanzhong 已提交
31

32 33 34 35 36 37
# NOTE(paddle-dev): All of these flags should be set before
# `import paddle`. Otherwise, it would not take any effect.
set_paddle_flags(
    FLAGS_eager_delete_tensor_gb=0,  # enable GC to save memory
)

38 39 40 41 42 43 44 45
from paddle import fluid

from ppdet.core.workspace import load_config, merge_config, create
from ppdet.modeling.model_input import create_feed
from ppdet.data.data_feed import create_reader

from ppdet.utils.eval_utils import parse_fetches
from ppdet.utils.cli import ArgsParser
W
wangguanzhong 已提交
46
from ppdet.utils.check import check_gpu, check_version
47 48 49 50 51 52 53 54 55 56 57 58 59 60 61
from ppdet.utils.visualizer import visualize_results
import ppdet.utils.checkpoint as checkpoint

import logging
FORMAT = '%(asctime)s-%(levelname)s: %(message)s'
logging.basicConfig(level=logging.INFO, format=FORMAT)
logger = logging.getLogger(__name__)


def get_save_image_name(output_dir, image_path):
    """
    Get save image name from source image path.
    """
    if not os.path.exists(output_dir):
        os.makedirs(output_dir)
K
Kaipeng Deng 已提交
62
    image_name = os.path.split(image_path)[-1]
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
    name, ext = os.path.splitext(image_name)
    return os.path.join(output_dir, "{}".format(name)) + ext


def get_test_images(infer_dir, infer_img):
    """
    Get image path list in TEST mode
    """
    assert infer_img is not None or infer_dir is not None, \
        "--infer_img or --infer_dir should be set"
    assert infer_img is None or os.path.isfile(infer_img), \
            "{} is not a file".format(infer_img)
    assert infer_dir is None or os.path.isdir(infer_dir), \
            "{} is not a directory".format(infer_dir)
    images = []

    # infer_img has a higher priority
    if infer_img and os.path.isfile(infer_img):
        images.append(infer_img)
        return images

    infer_dir = os.path.abspath(infer_dir)
    assert os.path.isdir(infer_dir), \
        "infer_dir {} is not a directory".format(infer_dir)
    exts = ['jpg', 'jpeg', 'png', 'bmp']
    exts += [ext.upper() for ext in exts]
    for ext in exts:
        images.extend(glob.glob('{}/*.{}'.format(infer_dir, ext)))

    assert len(images) > 0, "no image found in {}".format(infer_dir)
    logger.info("Found {} inference images in total.".format(len(images)))

    return images


def main():
    cfg = load_config(FLAGS.config)

    if 'architecture' in cfg:
        main_arch = cfg.architecture
    else:
        raise ValueError("'architecture' not specified in config file.")

    merge_config(FLAGS.opt)

108 109
    # check if set use_gpu=True in paddlepaddle cpu version
    check_gpu(cfg.use_gpu)
W
wangguanzhong 已提交
110 111
    # check if paddlepaddle version is satisfied
    check_version()
112

113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
    if 'test_feed' not in cfg:
        test_feed = create(main_arch + 'TestFeed')
    else:
        test_feed = create(cfg.test_feed)

    test_images = get_test_images(FLAGS.infer_dir, FLAGS.infer_img)
    test_feed.dataset.add_images(test_images)

    place = fluid.CUDAPlace(0) if cfg.use_gpu else fluid.CPUPlace()
    exe = fluid.Executor(place)

    model = create(main_arch)

    startup_prog = fluid.Program()
    infer_prog = fluid.Program()
    with fluid.program_guard(infer_prog, startup_prog):
        with fluid.unique_name.guard():
W
wangguanzhong 已提交
130
            loader, feed_vars = create_feed(test_feed, iterable=True)
131 132 133 134
            test_fetches = model.test(feed_vars)
    infer_prog = infer_prog.clone(True)

    reader = create_reader(test_feed)
W
wangguanzhong 已提交
135
    loader.set_sample_list_generator(reader, place)
136 137 138

    exe.run(startup_prog)
    if cfg.weights:
139
        checkpoint.load_params(exe, infer_prog, cfg.weights)
140 141

    # parse infer fetches
142
    assert cfg.metric in ['COCO', 'VOC', 'OID', 'WIDERFACE'], \
143
            "unknown metric type {}".format(cfg.metric)
144
    extra_keys = []
145
    if cfg['metric'] in ['COCO', 'OID']:
146
        extra_keys = ['im_info', 'im_id', 'im_shape']
147
    if cfg['metric'] == 'VOC' or cfg['metric'] == 'WIDERFACE':
148
        extra_keys = ['im_id', 'im_shape']
149 150 151 152 153
    keys, values, _ = parse_fetches(test_fetches, infer_prog, extra_keys)

    # parse dataset category
    if cfg.metric == 'COCO':
        from ppdet.utils.coco_eval import bbox2out, mask2out, get_category_info
154 155
    if cfg.metric == 'OID':
        from ppdet.utils.oid_eval import bbox2out, get_category_info
156 157
    if cfg.metric == "VOC":
        from ppdet.utils.voc_eval import bbox2out, get_category_info
158 159
    if cfg.metric == "WIDERFACE":
        from ppdet.utils.widerface_eval_utils import bbox2out, get_category_info
160 161 162 163 164 165 166

    anno_file = getattr(test_feed.dataset, 'annotation', None)
    with_background = getattr(test_feed, 'with_background', True)
    use_default_label = getattr(test_feed, 'use_default_label', False)
    clsid2catid, catid2name = get_category_info(anno_file, with_background,
                                                use_default_label)

167 168 169 170 171 172
    # whether output bbox is normalized in model output layer
    is_bbox_normalized = False
    if hasattr(model, 'is_bbox_normalized') and \
            callable(model.is_bbox_normalized):
        is_bbox_normalized = model.is_bbox_normalized()

173 174 175 176 177
    # use tb-paddle to log image
    if FLAGS.use_tb:
        from tb_paddle import SummaryWriter
        tb_writer = SummaryWriter(FLAGS.tb_log_dir)
        tb_image_step = 0
W
wangguanzhong 已提交
178
        tb_image_frame = 0  # each frame can display ten pictures at most. 
179

180
    imid2path = reader.imid2path
W
wangguanzhong 已提交
181
    for iter_id, data in enumerate(loader()):
182
        outs = exe.run(infer_prog,
W
wangguanzhong 已提交
183
                       feed=data,
184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
                       fetch_list=values,
                       return_numpy=False)
        res = {
            k: (np.array(v), v.recursive_sequence_lengths())
            for k, v in zip(keys, outs)
        }
        logger.info('Infer iter {}'.format(iter_id))

        bbox_results = None
        mask_results = None
        if 'bbox' in res:
            bbox_results = bbox2out([res], clsid2catid, is_bbox_normalized)
        if 'mask' in res:
            mask_results = mask2out([res], clsid2catid,
                                    model.mask_head.resolution)

        # visualize result
        im_ids = res['im_id'][0]
        for im_id in im_ids:
            image_path = imid2path[int(im_id)]
            image = Image.open(image_path).convert('RGB')
205 206 207 208 209 210 211 212 213 214

            # use tb-paddle to log original image           
            if FLAGS.use_tb:
                original_image_np = np.array(image)
                tb_writer.add_image(
                    "original/frame_{}".format(tb_image_frame),
                    original_image_np,
                    tb_image_step,
                    dataformats='HWC')

215
            image = visualize_results(image,
J
jerrywgz 已提交
216 217
                                      int(im_id), catid2name,
                                      FLAGS.draw_threshold, bbox_results,
218
                                      mask_results)
W
wangguanzhong 已提交
219

220 221 222 223 224 225 226 227 228 229 230 231 232
            # use tb-paddle to log image with bbox
            if FLAGS.use_tb:
                infer_image_np = np.array(image)
                tb_writer.add_image(
                    "bbox/frame_{}".format(tb_image_frame),
                    infer_image_np,
                    tb_image_step,
                    dataformats='HWC')
                tb_image_step += 1
                if tb_image_step % 10 == 0:
                    tb_image_step = 0
                    tb_image_frame += 1

233 234
            save_name = get_save_image_name(FLAGS.output_dir, image_path)
            logger.info("Detection bbox results save in {}".format(save_name))
J
jerrywgz 已提交
235
            image.save(save_name, quality=95)
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254


if __name__ == '__main__':
    parser = ArgsParser()
    parser.add_argument(
        "--infer_dir",
        type=str,
        default=None,
        help="Directory for images to perform inference on.")
    parser.add_argument(
        "--infer_img",
        type=str,
        default=None,
        help="Image path, has higher priority over --infer_dir")
    parser.add_argument(
        "--output_dir",
        type=str,
        default="output",
        help="Directory for storing the output visualization files.")
J
jerrywgz 已提交
255 256 257 258 259
    parser.add_argument(
        "--draw_threshold",
        type=float,
        default=0.5,
        help="Threshold to reserve the result for visualization.")
260 261 262 263 264 265 266 267 268 269
    parser.add_argument(
        "--use_tb",
        type=bool,
        default=False,
        help="whether to record the data to Tensorboard.")
    parser.add_argument(
        '--tb_log_dir',
        type=str,
        default="tb_log_dir/image",
        help='Tensorboard logging directory for image.')
270 271
    FLAGS = parser.parse_args()
    main()