Compute Library
 20.05
NEHOGMultiDetection.cpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2016-2020 ARM Limited.
3  *
4  * SPDX-License-Identifier: MIT
5  *
6  * Permission is hereby granted, free of charge, to any person obtaining a copy
7  * of this software and associated documentation files (the "Software"), to
8  * deal in the Software without restriction, including without limitation the
9  * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
10  * sell copies of the Software, and to permit persons to whom the Software is
11  * furnished to do so, subject to the following conditions:
12  *
13  * The above copyright notice and this permission notice shall be included in all
14  * copies or substantial portions of the Software.
15  *
16  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
19  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
21  * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
22  * SOFTWARE.
23  */
25 
26 #include "arm_compute/core/Error.h"
31 
32 using namespace arm_compute;
33 
34 NEHOGMultiDetection::NEHOGMultiDetection(std::shared_ptr<IMemoryManager> memory_manager) // NOLINT
35  : _memory_group(std::move(memory_manager)),
36  _gradient_kernel(),
37  _orient_bin_kernel(),
38  _block_norm_kernel(),
39  _hog_detect_kernel(),
40  _non_maxima_kernel(),
41  _hog_space(),
42  _hog_norm_space(),
43  _detection_windows(),
44  _mag(),
45  _phase(),
46  _non_maxima_suppression(false),
47  _num_orient_bin_kernel(0),
48  _num_block_norm_kernel(0),
49  _num_hog_detect_kernel(0)
50 {
51 }
52 
53 void NEHOGMultiDetection::configure(ITensor *input, const IMultiHOG *multi_hog, IDetectionWindowArray *detection_windows, const ISize2DArray *detection_window_strides, BorderMode border_mode,
54  uint8_t constant_border_value, float threshold, bool non_maxima_suppression, float min_distance)
55 {
58  ARM_COMPUTE_ERROR_ON(nullptr == detection_windows);
59  ARM_COMPUTE_ERROR_ON(detection_window_strides->num_values() != multi_hog->num_models());
60 
61  const size_t width = input->info()->dimension(Window::DimX);
62  const size_t height = input->info()->dimension(Window::DimY);
63  const TensorShape &shape_img = input->info()->tensor_shape();
64  const size_t num_models = multi_hog->num_models();
65  PhaseType phase_type = multi_hog->model(0)->info()->phase_type();
66 
67  size_t prev_num_bins = multi_hog->model(0)->info()->num_bins();
68  Size2D prev_cell_size = multi_hog->model(0)->info()->cell_size();
69  Size2D prev_block_size = multi_hog->model(0)->info()->block_size();
70  Size2D prev_block_stride = multi_hog->model(0)->info()->block_stride();
71 
72  /* Check if NEHOGOrientationBinningKernel and NEHOGBlockNormalizationKernel kernels can be skipped for a specific HOG data-object
73  *
74  * 1) NEHOGOrientationBinningKernel and NEHOGBlockNormalizationKernel are skipped if the cell size and the number of bins don't change.
75  * Since "multi_hog" is sorted,it is enough to check the HOG descriptors at level "ith" and level "(i-1)th
76  * 2) NEHOGBlockNormalizationKernel is skipped if the cell size, the number of bins and block size do not change.
77  * Since "multi_hog" is sorted,it is enough to check the HOG descriptors at level "ith" and level "(i-1)th
78  *
79  * @note Since the orientation binning and block normalization kernels can be skipped, we need to keep track of the input to process for each kernel
80  * with "input_orient_bin", "input_hog_detect" and "input_block_norm"
81  */
82  std::vector<size_t> input_orient_bin;
83  std::vector<size_t> input_hog_detect;
84  std::vector<std::pair<size_t, size_t>> input_block_norm;
85 
86  input_orient_bin.push_back(0);
87  input_hog_detect.push_back(0);
88  input_block_norm.emplace_back(0, 0);
89 
90  for(size_t i = 1; i < num_models; ++i)
91  {
92  size_t cur_num_bins = multi_hog->model(i)->info()->num_bins();
93  Size2D cur_cell_size = multi_hog->model(i)->info()->cell_size();
94  Size2D cur_block_size = multi_hog->model(i)->info()->block_size();
95  Size2D cur_block_stride = multi_hog->model(i)->info()->block_stride();
96 
97  if((cur_num_bins != prev_num_bins) || (cur_cell_size.width != prev_cell_size.width) || (cur_cell_size.height != prev_cell_size.height))
98  {
99  prev_num_bins = cur_num_bins;
100  prev_cell_size = cur_cell_size;
101  prev_block_size = cur_block_size;
102  prev_block_stride = cur_block_stride;
103 
104  // Compute orientation binning and block normalization kernels. Update input to process
105  input_orient_bin.push_back(i);
106  input_block_norm.emplace_back(i, input_orient_bin.size() - 1);
107  }
108  else if((cur_block_size.width != prev_block_size.width) || (cur_block_size.height != prev_block_size.height) || (cur_block_stride.width != prev_block_stride.width)
109  || (cur_block_stride.height != prev_block_stride.height))
110  {
111  prev_block_size = cur_block_size;
112  prev_block_stride = cur_block_stride;
113 
114  // Compute block normalization kernel. Update input to process
115  input_block_norm.emplace_back(i, input_orient_bin.size() - 1);
116  }
117 
118  // Update input to process for hog detector kernel
119  input_hog_detect.push_back(input_block_norm.size() - 1);
120  }
121 
122  _detection_windows = detection_windows;
123  _non_maxima_suppression = non_maxima_suppression;
124  _num_orient_bin_kernel = input_orient_bin.size(); // Number of NEHOGOrientationBinningKernel kernels to compute
125  _num_block_norm_kernel = input_block_norm.size(); // Number of NEHOGBlockNormalizationKernel kernels to compute
126  _num_hog_detect_kernel = input_hog_detect.size(); // Number of NEHOGDetector functions to compute
127 
128  _orient_bin_kernel.clear();
129  _block_norm_kernel.clear();
130  _hog_detect_kernel.clear();
131  _hog_space.clear();
132  _hog_norm_space.clear();
133 
134  _orient_bin_kernel.resize(_num_orient_bin_kernel);
135  _block_norm_kernel.resize(_num_block_norm_kernel);
136  _hog_detect_kernel.resize(_num_hog_detect_kernel);
137  _hog_space.resize(_num_orient_bin_kernel);
138  _hog_norm_space.resize(_num_block_norm_kernel);
139  _non_maxima_kernel = CPPDetectionWindowNonMaximaSuppressionKernel();
140 
141  // Allocate tensors for magnitude and phase
142  TensorInfo info_mag(shape_img, Format::S16);
143  _mag.allocator()->init(info_mag);
144 
145  TensorInfo info_phase(shape_img, Format::U8);
146  _phase.allocator()->init(info_phase);
147 
148  // Manage intermediate buffers
149  _memory_group.manage(&_mag);
150  _memory_group.manage(&_phase);
151 
152  // Initialise gradient kernel
153  _gradient_kernel.configure(input, &_mag, &_phase, phase_type, border_mode, constant_border_value);
154 
155  // Configure NETensor for the HOG space and orientation binning kernel
156  for(size_t i = 0; i < _num_orient_bin_kernel; ++i)
157  {
158  const size_t idx_multi_hog = input_orient_bin[i];
159 
160  // Get the corresponding cell size and number of bins
161  const Size2D &cell = multi_hog->model(idx_multi_hog)->info()->cell_size();
162  const size_t num_bins = multi_hog->model(idx_multi_hog)->info()->num_bins();
163 
164  // Calculate number of cells along the x and y directions for the hog_space
165  const size_t num_cells_x = width / cell.width;
166  const size_t num_cells_y = height / cell.height;
167 
168  // TensorShape of hog space
169  TensorShape shape_hog_space = input->info()->tensor_shape();
170  shape_hog_space.set(Window::DimX, num_cells_x);
171  shape_hog_space.set(Window::DimY, num_cells_y);
172 
173  // Allocate HOG space
174  TensorInfo info_space(shape_hog_space, num_bins, DataType::F32);
175  _hog_space[i].allocator()->init(info_space);
176 
177  // Manage intermediate buffers
178  _memory_group.manage(&_hog_space[i]);
179 
180  // Initialise orientation binning kernel
181  _orient_bin_kernel[i].configure(&_mag, &_phase, &_hog_space[i], multi_hog->model(idx_multi_hog)->info());
182  }
183 
184  // Allocate intermediate tensors
185  _mag.allocator()->allocate();
186  _phase.allocator()->allocate();
187 
188  // Configure NETensor for the normalized HOG space and block normalization kernel
189  for(size_t i = 0; i < _num_block_norm_kernel; ++i)
190  {
191  const size_t idx_multi_hog = input_block_norm[i].first;
192  const size_t idx_orient_bin = input_block_norm[i].second;
193 
194  // Allocate normalized HOG space
195  TensorInfo tensor_info(*(multi_hog->model(idx_multi_hog)->info()), width, height);
196  _hog_norm_space[i].allocator()->init(tensor_info);
197 
198  // Manage intermediate buffers
199  _memory_group.manage(&_hog_norm_space[i]);
200 
201  // Initialize block normalization kernel
202  _block_norm_kernel[i].configure(&_hog_space[idx_orient_bin], &_hog_norm_space[i], multi_hog->model(idx_multi_hog)->info());
203  }
204 
205  // Allocate intermediate tensors
206  for(size_t i = 0; i < _num_orient_bin_kernel; ++i)
207  {
208  _hog_space[i].allocator()->allocate();
209  }
210 
211  // Configure HOG detector kernel
212  for(size_t i = 0; i < _num_hog_detect_kernel; ++i)
213  {
214  const size_t idx_block_norm = input_hog_detect[i];
215 
216  _hog_detect_kernel[i].configure(&_hog_norm_space[idx_block_norm], multi_hog->model(i), detection_windows, detection_window_strides->at(i), threshold, i);
217  }
218 
219  // Configure non maxima suppression kernel
220  _non_maxima_kernel.configure(_detection_windows, min_distance);
221 
222  // Allocate intermediate tensors
223  for(size_t i = 0; i < _num_block_norm_kernel; ++i)
224  {
225  _hog_norm_space[i].allocator()->allocate();
226  }
227 }
228 
230 {
231  ARM_COMPUTE_ERROR_ON_MSG(_detection_windows == nullptr, "Unconfigured function");
232 
233  MemoryGroupResourceScope scope_mg(_memory_group);
234 
235  // Reset detection window
236  _detection_windows->clear();
237 
238  // Run gradient
239  _gradient_kernel.run();
240 
241  // Run orientation binning kernel
242  for(auto &kernel : _orient_bin_kernel)
243  {
245  }
246 
247  // Run block normalization kernel
248  for(auto &kernel : _block_norm_kernel)
249  {
251  }
252 
253  // Run HOG detector kernel
254  for(auto &kernel : _hog_detect_kernel)
255  {
256  kernel.run();
257  }
258 
259  // Run non-maxima suppression kernel if enabled
260  if(_non_maxima_suppression)
261  {
262  NEScheduler::get().schedule(&_non_maxima_kernel, Window::DimY);
263  }
264 }
BorderMode
Methods available to handle borders.
Definition: Types.h:264
size_t num_bins() const
The number of histogram bins for each cell.
Definition: HOGInfo.cpp:111
void clear()
Clear all the points from the array.
Definition: IArray.h:91
Shape of a tensor.
Definition: TensorShape.h:39
void init(const TensorAllocator &allocator, const Coordinates &coords, TensorInfo &sub_info)
Shares the same backing memory with another tensor allocator, while the tensor info might be differen...
const Size2D & cell_size() const
The cell size in pixels.
Definition: HOGInfo.cpp:91
PhaseType
Phase calculation type.
Definition: Types.h:409
1 channel, 1 U8 per channel
Array of type T.
Definition: IArray.h:40
1 channel, 1 F32 per channel
#define ARM_COMPUTE_ERROR_ON(cond)
If the condition is true then an error message is printed and an exception thrown.
Definition: Error.h:466
PhaseType phase_type() const
The type of PhaseType.
Definition: HOGInfo.cpp:126
Interface for NEON tensor.
Definition: ITensor.h:36
const Size2D & block_stride() const
The block stride in pixels.
Definition: HOGInfo.cpp:106
Copyright (c) 2017-2020 ARM Limited.
size_t height
Height of the image region or rectangle.
Definition: Size2D.h:90
#define ARM_COMPUTE_ERROR_ON_INVALID_MULTI_HOG(m)
Definition: Validate.h:925
void run() override
Run the kernels contained in the function.
TensorAllocator * allocator()
Return a pointer to the tensor's allocator.
Definition: Tensor.cpp:48
void configure(ITensor *input, ITensor *output_magnitude, ITensor *output_phase, PhaseType phase_type, BorderMode border_mode, uint8_t constant_border_value=0)
Initialise the function's source, destinations, phase type and border mode.
virtual size_t num_models() const =0
The number of HOG models stored.
void manage(IMemoryManageable *obj) override
Sets a object to be managed by the given memory group.
Definition: MemoryGroup.h:79
static constexpr size_t DimX
Alias for dimension 0 also known as X dimension.
Definition: Window.h:43
size_t num_values() const
Number of values currently stored in the array.
Definition: IArray.h:68
#define ARM_COMPUTE_ERROR_ON_MSG(cond, msg)
Definition: Error.h:456
const Size2D & block_size() const
The block size in pixels.
Definition: HOGInfo.cpp:96
void allocate() override
Allocate size specified by TensorInfo of CPU memory.
void configure(IDetectionWindowArray *input_output, float min_distance)
Initialise the kernel's input, output and the euclidean minimum distance.
void run() override
Run the kernels contained in the function.
1 channel, 1 S16 per channel
CPP kernel to perform in-place computation of euclidean distance on IDetectionWindowArray.
#define ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c,...)
Definition: Validate.h:790
void configure(ITensor *input, const IMultiHOG *multi_hog, IDetectionWindowArray *detection_windows, const ISize2DArray *detection_window_strides, BorderMode border_mode, uint8_t constant_border_value=0, float threshold=0.0f, bool non_maxima_suppression=false, float min_distance=1.0f)
Initialise the function's source, destination, detection window strides, border mode,...
static constexpr size_t DimY
Alias for dimension 1 also known as Y dimension.
Definition: Window.h:45
virtual IHOG * model(size_t index)=0
Return a pointer to the requested HOG model.
Memory group resources scope handling class.
Definition: IMemoryGroup.h:82
virtual void schedule(ICPPKernel *kernel, const Hints &hints)=0
Runs the kernel in the same thread as the caller synchronously.
size_t width
Width of the image region or rectangle.
Definition: Size2D.h:89
Interface for storing multiple HOG data-objects.
Definition: IMultiHOG.h:34
SimpleTensor< T > non_maxima_suppression(const SimpleTensor< T > &src, BorderMode border_mode, T constant_border_value)
Class for specifying the size of an image or rectangle.
Definition: Size2D.h:34
TensorShape & set(size_t dimension, size_t value, bool apply_dim_correction=true)
Accessor to set the value of one of the dimensions.
Definition: TensorShape.h:78
Store the tensor's metadata.
Definition: TensorInfo.h:45
NEHOGMultiDetection(std::shared_ptr< IMemoryManager > memory_manager=nullptr)
Default constructor.
virtual const HOGInfo * info() const =0
Interface to be implemented by the child class to return the HOG's metadata.
SimpleTensor< T > threshold(const SimpleTensor< T > &src, T threshold, T false_value, T true_value, ThresholdType type, T upper)
Definition: Threshold.cpp:35
virtual T & at(size_t index) const
Reference to the element of the array located at the given index.
Definition: IArray.h:117
static IScheduler & get()
Access the scheduler singleton.
Definition: Scheduler.cpp:95