Skip to content

O
opencv
Commit 4fb086d6 authored by Dmitry Kurtaev's avatar Dmitry Kurtaev
Browse files
Options

MobileNet-SSD v1 from TensorFlow with shared convolution weights

parent 67f79aab
Hide whitespace changes
Inline Side-by-side
Showing with 103 additions and 46 deletions
modules/dnn/src/tensorflow/tf_importer.cpp
View file @ 4fb086d6
......@@ -716,6 +716,8 @@ void TFImporter::populateNet(Net dstNet)
// find all Const layers for params
std::map<String, int> value_id;
// A map with constant blobs which are shared between multiple layers.
std::map<String, Mat> sharedWeights;
addConstNodes(netBin, value_id, layers_to_ignore);
addConstNodes(netTxt, value_id, layers_to_ignore);
......@@ -805,52 +807,65 @@ void TFImporter::populateNet(Net dstNet)
}
}
const tensorflow::TensorProto& kernelTensor = getConstBlob(layer, value_id);
kernelFromTensor(kernelTensor, layerParams.blobs[0]);
releaseTensor(const_cast<tensorflow::TensorProto*>(&kernelTensor));
int* kshape = layerParams.blobs[0].size.p;
const int outCh = kshape[0];
const int inCh = kshape[1];
const int height = kshape[2];
const int width = kshape[3];
if (type == "DepthwiseConv2dNative")
int kernelTensorInpId = -1;
const tensorflow::TensorProto& kernelTensor = getConstBlob(layer, value_id, -1, &kernelTensorInpId);
const String kernelTensorName = layer.input(kernelTensorInpId);
std::map<String, Mat>::iterator sharedWeightsIt = sharedWeights.find(kernelTensorName);
if (sharedWeightsIt == sharedWeights.end())
{
CV_Assert(!locPredTransposed);
const int chMultiplier = kshape[0];
Mat copy = layerParams.blobs[0].clone();
float* src = (float*)copy.data;
float* dst = (float*)layerParams.blobs[0].data;
for (int i = 0; i < chMultiplier; ++i)
for (int j = 0; j < inCh; ++j)
for (int s = 0; s < height * width; ++s)
{
int src_i = (i * inCh + j) * height * width + s;
int dst_i = (j * chMultiplier + i) * height* width + s;
dst[dst_i] = src[src_i];
}
// TODO Use reshape instead
kshape[0] = inCh * chMultiplier;
kshape[1] = 1;
size_t* kstep = layerParams.blobs[0].step.p;
kstep[0] = kstep[1]; // fix steps too
}
layerParams.set("kernel_h", height);
layerParams.set("kernel_w", width);
layerParams.set("num_output", outCh);
kernelFromTensor(kernelTensor, layerParams.blobs[0]);
releaseTensor(const_cast<tensorflow::TensorProto*>(&kernelTensor));
int* kshape = layerParams.blobs[0].size.p;
const int outCh = kshape[0];
const int inCh = kshape[1];
const int height = kshape[2];
const int width = kshape[3];
if (type == "DepthwiseConv2dNative")
{
CV_Assert(!locPredTransposed);
const int chMultiplier = kshape[0];
Mat copy = layerParams.blobs[0].clone();
float* src = (float*)copy.data;
float* dst = (float*)layerParams.blobs[0].data;
for (int i = 0; i < chMultiplier; ++i)
for (int j = 0; j < inCh; ++j)
for (int s = 0; s < height * width; ++s)
{
int src_i = (i * inCh + j) * height * width + s;
int dst_i = (j * chMultiplier + i) * height* width + s;
dst[dst_i] = src[src_i];
}
// TODO Use reshape instead
kshape[0] = inCh * chMultiplier;
kshape[1] = 1;
size_t* kstep = layerParams.blobs[0].step.p;
kstep[0] = kstep[1]; // fix steps too
}
// Shuffle output channels from yxYX to xyXY.
if (locPredTransposed)
{
const int slice = height * width * inCh;
for (int i = 0; i < outCh; i += 2)
// Shuffle output channels from yxYX to xyXY.
if (locPredTransposed)
{
cv::Mat src(1, slice, CV_32F, layerParams.blobs[0].ptr<float>(i));
cv::Mat dst(1, slice, CV_32F, layerParams.blobs[0].ptr<float>(i + 1));
std::swap_ranges(src.begin<float>(), src.end<float>(), dst.begin<float>());
const int slice = height * width * inCh;
for (int i = 0; i < outCh; i += 2)
{
cv::Mat src(1, slice, CV_32F, layerParams.blobs[0].ptr<float>(i));
cv::Mat dst(1, slice, CV_32F, layerParams.blobs[0].ptr<float>(i + 1));
std::swap_ranges(src.begin<float>(), src.end<float>(), dst.begin<float>());
}
}
sharedWeights[kernelTensorName] = layerParams.blobs[0];
}
else
{
layerParams.blobs[0] = sharedWeightsIt->second;
}
layerParams.set("kernel_h", layerParams.blobs[0].size[2]);
layerParams.set("kernel_w", layerParams.blobs[0].size[3]);
layerParams.set("num_output", layerParams.blobs[0].size[0]);
setStrides(layerParams, layer);
setPadding(layerParams, layer);
......
modules/dnn/test/test_tf_importer.cpp
View file @ 4fb086d6
......@@ -343,6 +343,26 @@ TEST_P(Test_TensorFlow_nets, Inception_v2_Faster_RCNN)
normAssertDetections(ref, out, "", 0.3);
}
TEST_P(Test_TensorFlow_nets, MobileNet_v1_SSD_PPN)
{
checkBackend();
std::string proto = findDataFile("dnn/ssd_mobilenet_v1_ppn_coco.pbtxt", false);
std::string model = findDataFile("dnn/ssd_mobilenet_v1_ppn_coco.pb", false);
Net net = readNetFromTensorflow(model, proto);
Mat img = imread(findDataFile("dnn/dog416.png", false));
Mat ref = blobFromNPY(findDataFile("dnn/tensorflow/ssd_mobilenet_v1_ppn_coco.detection_out.npy", false));
Mat blob = blobFromImage(img, 1.0f / 127.5, Size(300, 300), Scalar(127.5, 127.5, 127.5), true, false);
net.setPreferableBackend(backend);
net.setPreferableTarget(target);
net.setInput(blob);
Mat out = net.forward();
double scoreDiff = (target == DNN_TARGET_OPENCL_FP16 || target == DNN_TARGET_MYRIAD) ? 0.006 : default_l1;
normAssertDetections(ref, out, "", 0.4, scoreDiff, default_lInf);
}
TEST_P(Test_TensorFlow_nets, opencv_face_detector_uint8)
{
checkBackend();
......
samples/dnn/tf_text_graph_ssd.py
View file @ 4fb086d6
......@@ -29,6 +29,11 @@ parser.add_argument('--aspect_ratios', default=[1.0, 2.0, 0.5, 3.0, 0.333], type
help='Hyper-parameter of ssd_anchor_generator from config file.')
parser.add_argument('--image_width', default=300, type=int, help='Training images width.')
parser.add_argument('--image_height', default=300, type=int, help='Training images height.')
parser.add_argument('--not_reduce_boxes_in_lowest_layer', default=False, action='store_true',
help='A boolean to indicate whether the fixed 3 boxes per '
'location is used in the lowest achors generation layer.')
parser.add_argument('--box_predictor', default='convolutional', type=str,
choices=['convolutional', 'weight_shared_convolutional'])
args = parser.parse_args()
# Nodes that should be kept.
......@@ -194,12 +199,18 @@ def addConcatNode(name, inputs, axisNodeName):
addConstNode('concat/axis_flatten', [-1])
addConstNode('PriorBox/concat/axis', [-2])
for label in ['ClassPredictor', 'BoxEncodingPredictor']:
for label in ['ClassPredictor', 'BoxEncodingPredictor' if args.box_predictor is 'convolutional' else 'BoxPredictor']:
concatInputs = []
for i in range(args.num_layers):
# Flatten predictions
flatten = NodeDef()
inpName = 'BoxPredictor_%d/%s/BiasAdd' % (i, label)
if args.box_predictor is 'convolutional':
inpName = 'BoxPredictor_%d/%s/BiasAdd' % (i, label)
else:
if i == 0:
inpName = 'WeightSharedConvolutionalBoxPredictor/%s/BiasAdd' % label
else:
inpName = 'WeightSharedConvolutionalBoxPredictor_%d/%s/BiasAdd' % (i, label)
flatten.input.append(inpName)
flatten.name = inpName + '/Flatten'
flatten.op = 'Flatten'
......@@ -210,7 +221,9 @@ for label in ['ClassPredictor', 'BoxEncodingPredictor']:
idx = 0
for node in graph_def.node:
if node.name == ('BoxPredictor_%d/BoxEncodingPredictor/Conv2D' % idx):
if node.name == ('BoxPredictor_%d/BoxEncodingPredictor/Conv2D' % idx) or \
node.name == ('WeightSharedConvolutionalBoxPredictor_%d/BoxPredictor/Conv2D' % idx) or \
node.name == 'WeightSharedConvolutionalBoxPredictor/BoxPredictor/Conv2D':
text_format.Merge('b: true', node.attr["loc_pred_transposed"])
idx += 1
assert(idx == args.num_layers)
......@@ -224,13 +237,19 @@ for i in range(args.num_layers):
priorBox = NodeDef()
priorBox.name = 'PriorBox_%d' % i
priorBox.op = 'PriorBox'
priorBox.input.append('BoxPredictor_%d/BoxEncodingPredictor/BiasAdd' % i)
if args.box_predictor is 'convolutional':
priorBox.input.append('BoxPredictor_%d/BoxEncodingPredictor/BiasAdd' % i)
else:
if i == 0:
priorBox.input.append('WeightSharedConvolutionalBoxPredictor/BoxPredictor/Conv2D')
else:
priorBox.input.append('WeightSharedConvolutionalBoxPredictor_%d/BoxPredictor/BiasAdd' % i)
priorBox.input.append(graph_def.node[0].name) # image_tensor
text_format.Merge('b: false', priorBox.attr["flip"])
text_format.Merge('b: false', priorBox.attr["clip"])
if i == 0:
if i == 0 and not args.not_reduce_boxes_in_lowest_layer:
widths = [0.1, args.min_scale * sqrt(2.0), args.min_scale * sqrt(0.5)]
heights = [0.1, args.min_scale / sqrt(2.0), args.min_scale / sqrt(0.5)]
else:
......@@ -261,7 +280,10 @@ detectionOut = NodeDef()
detectionOut.name = 'detection_out'
detectionOut.op = 'DetectionOutput'
detectionOut.input.append('BoxEncodingPredictor/concat')
if args.box_predictor == 'convolutional':
detectionOut.input.append('BoxEncodingPredictor/concat')
else:
detectionOut.input.append('BoxPredictor/concat')
detectionOut.input.append(sigmoid.name)
detectionOut.input.append('PriorBox/concat')
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment