src/layer/yolodetectionoutput.cpp

// Tencent is pleased to support the open source community by making ncnn available.
//
// Copyright (C) 2018 THL A29 Limited, a Tencent company. All rights reserved.
//
// Licensed under the BSD 3-Clause License (the "License"); you may not use this file except
// in compliance with the License. You may obtain a copy of the License at
//
// https://opensource.org/licenses/BSD-3-Clause
//
// 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.

#include "yolodetectionoutput.h"

#include "layer_type.h"

#include <math.h>

namespace ncnn {

YoloDetectionOutput::YoloDetectionOutput()
{
    one_blob_only = false;
    support_inplace = true;
}

int YoloDetectionOutput::load_param(const ParamDict& pd)
{
    num_class = pd.get(0, 20);
    num_box = pd.get(1, 5);
    confidence_threshold = pd.get(2, 0.01f);
    nms_threshold = pd.get(3, 0.45f);
    biases = pd.get(4, Mat());

    return 0;
}

int YoloDetectionOutput::create_pipeline(const Option& opt)
{
    {
        softmax = ncnn::create_layer(ncnn::LayerType::Softmax);

        ncnn::ParamDict pd;
        pd.set(0, 0); // axis

        softmax->load_param(pd);

        softmax->create_pipeline(opt);
    }

    return 0;
}

int YoloDetectionOutput::destroy_pipeline(const Option& opt)
{
    if (softmax)
    {
        softmax->destroy_pipeline(opt);
        delete softmax;
        softmax = 0;
    }

    return 0;
}

struct BBoxRect
{
    float xmin;
    float ymin;
    float xmax;
    float ymax;
    int label;
};

static inline float intersection_area(const BBoxRect& a, const BBoxRect& b)
{
    if (a.xmin > b.xmax || a.xmax < b.xmin || a.ymin > b.ymax || a.ymax < b.ymin)
    {
        // no intersection
        return 0.f;
    }

    float inter_width = std::min(a.xmax, b.xmax) - std::max(a.xmin, b.xmin);
    float inter_height = std::min(a.ymax, b.ymax) - std::max(a.ymin, b.ymin);

    return inter_width * inter_height;
}

template<typename T>
static void qsort_descent_inplace(std::vector<T>& datas, std::vector<float>& scores, int left, int right)
{
    int i = left;
    int j = right;
    float p = scores[(left + right) / 2];

    while (i <= j)
    {
        while (scores[i] > p)
            i++;

        while (scores[j] < p)
            j--;

        if (i <= j)
        {
            // swap
            std::swap(datas[i], datas[j]);
            std::swap(scores[i], scores[j]);

            i++;
            j--;
        }
    }

    if (left < j)
        qsort_descent_inplace(datas, scores, left, j);

    if (i < right)
        qsort_descent_inplace(datas, scores, i, right);
}

template<typename T>
static void qsort_descent_inplace(std::vector<T>& datas, std::vector<float>& scores)
{
    if (datas.empty() || scores.empty())
        return;

    qsort_descent_inplace(datas, scores, 0, static_cast<int>(scores.size() - 1));
}

static void nms_sorted_bboxes(const std::vector<BBoxRect>& bboxes, std::vector<size_t>& picked, float nms_threshold)
{
    picked.clear();

    const size_t n = bboxes.size();

    std::vector<float> areas(n);
    for (size_t i = 0; i < n; i++)
    {
        const BBoxRect& r = bboxes[i];

        float width = r.xmax - r.xmin;
        float height = r.ymax - r.ymin;

        areas[i] = width * height;
    }

    for (size_t i = 0; i < n; i++)
    {
        const BBoxRect& a = bboxes[i];

        int keep = 1;
        for (int j = 0; j < (int)picked.size(); j++)
        {
            const BBoxRect& b = bboxes[picked[j]];

            // intersection over union
            float inter_area = intersection_area(a, b);
            float union_area = areas[i] + areas[picked[j]] - inter_area;
            //             float IoU = inter_area / union_area
            if (inter_area / union_area > nms_threshold)
                keep = 0;
        }

        if (keep)
            picked.push_back(i);
    }
}

static inline float sigmoid(float x)
{
    return static_cast<float>(1.f / (1.f + exp(-x)));
}

int YoloDetectionOutput::forward_inplace(std::vector<Mat>& bottom_top_blobs, const Option& opt) const
{
    // gather all box
    std::vector<BBoxRect> all_bbox_rects;
    std::vector<float> all_bbox_scores;

    for (size_t b = 0; b < bottom_top_blobs.size(); b++)
    {
        Mat& bottom_top_blob = bottom_top_blobs[b];

        int w = bottom_top_blob.w;
        int h = bottom_top_blob.h;
        int channels = bottom_top_blob.c;

        const int channels_per_box = channels / num_box;

        // anchor coord + box score + num_class
        if (channels_per_box != 4 + 1 + num_class)
            return -1;

        std::vector<std::vector<BBoxRect> > all_box_bbox_rects;
        std::vector<std::vector<float> > all_box_bbox_scores;
        all_box_bbox_rects.resize(num_box);
        all_box_bbox_scores.resize(num_box);

        #pragma omp parallel for num_threads(opt.num_threads)
        for (int pp = 0; pp < num_box; pp++)
        {
            int p = pp * channels_per_box;

            const float bias_w = biases[pp * 2];
            const float bias_h = biases[pp * 2 + 1];

            const float* xptr = bottom_top_blob.channel(p);
            const float* yptr = bottom_top_blob.channel(p + 1);
            const float* wptr = bottom_top_blob.channel(p + 2);
            const float* hptr = bottom_top_blob.channel(p + 3);

            const float* box_score_ptr = bottom_top_blob.channel(p + 4);

            // softmax class scores
            Mat scores = bottom_top_blob.channel_range(p + 5, num_class);
            softmax->forward_inplace(scores, opt);

            for (int i = 0; i < h; i++)
            {
                for (int j = 0; j < w; j++)
                {
                    // region box
                    float bbox_cx = (j + sigmoid(xptr[0])) / w;
                    float bbox_cy = (i + sigmoid(yptr[0])) / h;
                    float bbox_w = static_cast<float>(exp(wptr[0]) * bias_w / w);
                    float bbox_h = static_cast<float>(exp(hptr[0]) * bias_h / h);

                    float bbox_xmin = bbox_cx - bbox_w * 0.5f;
                    float bbox_ymin = bbox_cy - bbox_h * 0.5f;
                    float bbox_xmax = bbox_cx + bbox_w * 0.5f;
                    float bbox_ymax = bbox_cy + bbox_h * 0.5f;

                    // box score
                    float box_score = sigmoid(box_score_ptr[0]);

                    // find class index with max class score
                    int class_index = 0;
                    float class_score = 0.f;
                    for (int q = 0; q < num_class; q++)
                    {
                        float score = scores.channel(q).row(i)[j];
                        if (score > class_score)
                        {
                            class_index = q;
                            class_score = score;
                        }
                    }

                    //                 NCNN_LOGE("%d %f %f", class_index, box_score, class_score);

                    float confidence = box_score * class_score;
                    if (confidence >= confidence_threshold)
                    {
                        BBoxRect c = {bbox_xmin, bbox_ymin, bbox_xmax, bbox_ymax, class_index};
                        all_box_bbox_rects[pp].push_back(c);
                        all_box_bbox_scores[pp].push_back(confidence);
                    }

                    xptr++;
                    yptr++;
                    wptr++;
                    hptr++;

                    box_score_ptr++;
                }
            }
        }

        for (int i = 0; i < num_box; i++)
        {
            const std::vector<BBoxRect>& box_bbox_rects = all_box_bbox_rects[i];
            const std::vector<float>& box_bbox_scores = all_box_bbox_scores[i];

            all_bbox_rects.insert(all_bbox_rects.end(), box_bbox_rects.begin(), box_bbox_rects.end());
            all_bbox_scores.insert(all_bbox_scores.end(), box_bbox_scores.begin(), box_bbox_scores.end());
        }
    }

    // global sort inplace
    qsort_descent_inplace(all_bbox_rects, all_bbox_scores);

    // apply nms
    std::vector<size_t> picked;
    nms_sorted_bboxes(all_bbox_rects, picked, nms_threshold);

    // select
    std::vector<BBoxRect> bbox_rects;
    std::vector<float> bbox_scores;

    for (size_t i = 0; i < picked.size(); i++)
    {
        size_t z = picked[i];
        bbox_rects.push_back(all_bbox_rects[z]);
        bbox_scores.push_back(all_bbox_scores[z]);
    }

    // fill result
    int num_detected = static_cast<int>(bbox_rects.size());
    if (num_detected == 0)
        return 0;

    Mat& top_blob = bottom_top_blobs[0];
    top_blob.create(6, num_detected, 4u, opt.blob_allocator);
    if (top_blob.empty())
        return -100;

    for (int i = 0; i < num_detected; i++)
    {
        const BBoxRect& r = bbox_rects[i];
        float score = bbox_scores[i];
        float* outptr = top_blob.row(i);

        outptr[0] = static_cast<float>(r.label + 1); // +1 for prepend background class
        outptr[1] = score;
        outptr[2] = r.xmin;
        outptr[3] = r.ymin;
        outptr[4] = r.xmax;
        outptr[5] = r.ymax;
    }

    return 0;
}

} // namespace ncnn