AlexeyAB · ZouJiu1 · Aug 8, 2023 · Aug 8, 2023 · Aug 8, 2023 · Aug 8, 2023
diff --git a/include/darknet.h b/include/darknet.h
@@ -101,7 +101,6 @@ typedef struct tree {
     int *group_offset;
 } tree;
 
-
 // activations.h
 typedef enum {
     LOGISTIC, RELU, RELU6, RELIE, LINEAR, RAMP, TANH, PLSE, REVLEAKY, LEAKY, ELU, LOGGY, STAIR, HARDTAN, LHTAN, SELU, GELU, SWISH, MISH, HARD_MISH, NORM_CHAN, NORM_CHAN_SOFTMAX, NORM_CHAN_SOFTMAX_MAXVAL
@@ -912,6 +911,18 @@ typedef struct detection{
     int track_id;
 } detection;
 
+// multi_label_one_anchor_choose_maxiou
+typedef struct mloam { 
+    box truth;
+    float best_iou;
+    int x;
+    int y;
+    int track_id;
+    int best_n;
+    int mask_n;
+    int t;
+} mloam;
+
 // network.c -batch inference
 typedef struct det_num_pair {
     int num;

diff --git a/src/yolo_layer.c b/src/yolo_layer.c
@@ -4,6 +4,7 @@
 #include "box.h"
 #include "dark_cuda.h"
 #include "utils.h"
+#include "darknet.h"
 
 #include <math.h>
 #include <stdio.h>
@@ -382,6 +383,7 @@ typedef struct train_yolo_args {
     float tot_iou_loss;
     int count;
     int class_count;
+    mloam* mloam_ptr;
 } train_yolo_args;
 
 void *process_batch(void* ptr)
@@ -502,6 +504,8 @@ void *process_batch(void* ptr)
                 }
             }
         }
+        mloam* mloam_ptr = args->mloam_ptr;   // multi_label_one_anchor_choose_maxiou
+        int number = 0;
         for (t = 0; t < l.max_boxes; ++t) {
             box truth = float_to_box_stride(state.truth + t * l.truth_size + b * l.truths, 1);
             if (!truth.x) break;  // continue;
@@ -534,54 +538,29 @@ void *process_batch(void* ptr)
 
             int mask_n = int_index(l.mask, best_n, l.n);
             if (mask_n >= 0) {
-                int class_id = state.truth[t * l.truth_size + b * l.truths + 4];
-                if (l.map) class_id = l.map[class_id];
-
-                int box_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 0);
-                const float class_multiplier = (l.classes_multipliers) ? l.classes_multipliers[class_id] : 1.0f;
-                ious all_ious = delta_yolo_box(truth, l.output, l.biases, best_n, box_index, i, j, l.w, l.h, state.net.w, state.net.h, l.delta, (2 - truth.w * truth.h), l.w * l.h, l.iou_normalizer * class_multiplier, l.iou_loss, 1, l.max_delta, state.net.rewritten_bbox, l.new_coords);
-                (*state.net.total_bbox)++;
-
-                const int truth_in_index = t * l.truth_size + b * l.truths + 5;
-                const int track_id = state.truth[truth_in_index];
-                const int truth_out_index = b * l.n * l.w * l.h + mask_n * l.w * l.h + j * l.w + i;
-                l.labels[truth_out_index] = track_id;
-                l.class_ids[truth_out_index] = class_id;
-                //printf(" track_id = %d, t = %d, b = %d, truth_in_index = %d, truth_out_index = %d \n", track_id, t, b, truth_in_index, truth_out_index);
-
-                // range is 0 <= 1
-                args->tot_iou += all_ious.iou;
-                args->tot_iou_loss += 1 - all_ious.iou;
-                // range is -1 <= giou <= 1
-                tot_giou += all_ious.giou;
-                args->tot_giou_loss += 1 - all_ious.giou;
-
-                tot_diou += all_ious.diou;
-                tot_diou_loss += 1 - all_ious.diou;
-
-                tot_ciou += all_ious.ciou;
-                tot_ciou_loss += 1 - all_ious.ciou;
-
-                int obj_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 4);
-                avg_obj += l.output[obj_index];
-                if (l.objectness_smooth) {
-                    float delta_obj = class_multiplier * l.obj_normalizer * (1 - l.output[obj_index]);
-                    if (l.delta[obj_index] == 0) l.delta[obj_index] = delta_obj;
+                if(number==0) {
+                    mloam choose = {truth, best_iou, i, j, 1, best_n, mask_n, t};
+                    mloam_ptr[number++] = choose;
+                }
+                else {
+                    int markable = 666;
+                    for(int mi = 0; mi < number; mi++) {
+                        mloam mp = mloam_ptr[mi];
+                        if(mp.best_n==best_n && mp.mask_n == mask_n && mp.x==i && mp.y==j) {
+                            markable = -666;
+                            if(mp.best_iou < best_iou) {
+                                mloam choose = {truth, best_iou, i, j, 1, best_n, mask_n, t};
+                                mloam_ptr[mi] = choose;
+                                break;
+                            }
+                        }
+                    }
+                    if(markable > 0) {
+                        mloam choose = {truth, best_iou, i, j, 1, best_n, mask_n, t};
+                        mloam_ptr[number++] = choose;
+                    }
                 }
-                else l.delta[obj_index] = class_multiplier * l.obj_normalizer * (1 - l.output[obj_index]);
-
-                int class_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 4 + 1);
-                delta_yolo_class(l.output, l.delta, class_index, class_id, l.classes, l.w * l.h, &avg_cat, l.focal_loss, l.label_smooth_eps, l.classes_multipliers, l.cls_normalizer);
-
-                //printf(" label: class_id = %d, truth.x = %f, truth.y = %f, truth.w = %f, truth.h = %f \n", class_id, truth.x, truth.y, truth.w, truth.h);
-                //printf(" mask_n = %d, l.output[obj_index] = %f, l.output[class_index + class_id] = %f \n\n", mask_n, l.output[obj_index], l.output[class_index + class_id]);
-
-                ++(args->count);
-                ++(args->class_count);
-                if (all_ious.iou > .5) recall += 1;
-                if (all_ious.iou > .75) recall75 += 1;
             }
-
             // iou_thresh
             for (n = 0; n < l.total; ++n) {
                 int mask_n = int_index(l.mask, n, l.n);
@@ -593,45 +572,89 @@ void *process_batch(void* ptr)
                     // iou, n
 
                     if (iou > l.iou_thresh) {
-                        int class_id = state.truth[t * l.truth_size + b * l.truths + 4];
-                        if (l.map) class_id = l.map[class_id];
-
-                        int box_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 0);
-                        const float class_multiplier = (l.classes_multipliers) ? l.classes_multipliers[class_id] : 1.0f;
-                        ious all_ious = delta_yolo_box(truth, l.output, l.biases, n, box_index, i, j, l.w, l.h, state.net.w, state.net.h, l.delta, (2 - truth.w * truth.h), l.w * l.h, l.iou_normalizer * class_multiplier, l.iou_loss, 1, l.max_delta, state.net.rewritten_bbox, l.new_coords);
-                        (*state.net.total_bbox)++;
+                        if(number==0) {
+                            mloam choose = {truth, iou, i, j, -1, n, mask_n, t};
+                            mloam_ptr[number++] = choose;
+                        }
+                        else {
+                            int markable = 666;
+                            for(int mi = 0; mi < number; mi++) {
+                                mloam mp = mloam_ptr[mi];
+                                if(mp.best_n==n && mp.mask_n == mask_n && mp.x==i && mp.y==j) {
+                                    markable = -666;
+                                    if(mp.best_iou < iou) {
+                                        mloam choose = {truth, iou, i, j, -1, n, mask_n, t};
+                                        mloam_ptr[mi] = choose;
+                                        break;
+                                    }
+                                }
+                            }
+                            if(markable > 0) {
+                                mloam choose = {truth, iou, i, j, -1, n, mask_n, t};
+                                mloam_ptr[number++] = choose;
+                            }
+                        }
+                    }
+                }
+            }
+        }
+        for(int ni = 0; ni < number; ni++) {
+            mloam mp   = mloam_ptr[ni];
+            box truth      = mp.truth;
+            int i          = mp.x;
+            int j          = mp.y;
+            int track_id   = mp.track_id;
+            int best_n     = mp.best_n;
+            int mask_n     = mp.mask_n;
+            int t          = mp.t;
 
-                        // range is 0 <= 1
-                        args->tot_iou += all_ious.iou;
-                        args->tot_iou_loss += 1 - all_ious.iou;
-                        // range is -1 <= giou <= 1
-                        tot_giou += all_ious.giou;
-                        args->tot_giou_loss += 1 - all_ious.giou;
+            int class_id = state.truth[t * l.truth_size + b * l.truths + 4];
+            if (l.map) class_id = l.map[class_id];
 
-                        tot_diou += all_ious.diou;
-                        tot_diou_loss += 1 - all_ious.diou;
+            int box_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 0);
+            const float class_multiplier = (l.classes_multipliers) ? l.classes_multipliers[class_id] : 1.0f;
+            ious all_ious = delta_yolo_box(truth, l.output, l.biases, best_n, box_index, i, j, l.w, l.h, state.net.w, state.net.h, l.delta, (2 - truth.w * truth.h), l.w * l.h, l.iou_normalizer * class_multiplier, l.iou_loss, 1, l.max_delta, state.net.rewritten_bbox, l.new_coords);
+            (*state.net.total_bbox)++;
 
-                        tot_ciou += all_ious.ciou;
-                        tot_ciou_loss += 1 - all_ious.ciou;
+            if(track_id > 0) {
+                const int truth_in_index = t * l.truth_size + b * l.truths + 5;
+                const int track_id = state.truth[truth_in_index];
+                const int truth_out_index = b * l.n * l.w * l.h + mask_n * l.w * l.h + j * l.w + i;
+                l.labels[truth_out_index] = track_id;
+                l.class_ids[truth_out_index] = class_id;
+                //printf(" track_id = %d, t = %d, b = %d, truth_in_index = %d, truth_out_index = %d \n", track_id, t, b, truth_in_index, truth_out_index);
+            }
+            // range is 0 <= 1
+            args->tot_iou += all_ious.iou;
+            args->tot_iou_loss += 1 - all_ious.iou;
+            // range is -1 <= giou <= 1
+            tot_giou += all_ious.giou;
+            args->tot_giou_loss += 1 - all_ious.giou;
+
+            tot_diou += all_ious.diou;
+            tot_diou_loss += 1 - all_ious.diou;
+
+            tot_ciou += all_ious.ciou;
+            tot_ciou_loss += 1 - all_ious.ciou;
+
+            int obj_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 4);
+            avg_obj += l.output[obj_index];
+            if (l.objectness_smooth) {
+                float delta_obj = class_multiplier * l.obj_normalizer * (1 - l.output[obj_index]);
+                if (l.delta[obj_index] == 0) l.delta[obj_index] = delta_obj;
+            }
+            else l.delta[obj_index] = class_multiplier * l.obj_normalizer * (1 - l.output[obj_index]);
 
-                        int obj_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 4);
-                        avg_obj += l.output[obj_index];
-                        if (l.objectness_smooth) {
-                            float delta_obj = class_multiplier * l.obj_normalizer * (1 - l.output[obj_index]);
-                            if (l.delta[obj_index] == 0) l.delta[obj_index] = delta_obj;
-                        }
-                        else l.delta[obj_index] = class_multiplier * l.obj_normalizer * (1 - l.output[obj_index]);
+            int class_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 4 + 1);
+            delta_yolo_class(l.output, l.delta, class_index, class_id, l.classes, l.w * l.h, &avg_cat, l.focal_loss, l.label_smooth_eps, l.classes_multipliers, l.cls_normalizer);
 
-                        int class_index = entry_index(l, b, mask_n * l.w * l.h + j * l.w + i, 4 + 1);
-                        delta_yolo_class(l.output, l.delta, class_index, class_id, l.classes, l.w * l.h, &avg_cat, l.focal_loss, l.label_smooth_eps, l.classes_multipliers, l.cls_normalizer);
+            //printf(" label: class_id = %d, truth.x = %f, truth.y = %f, truth.w = %f, truth.h = %f \n", class_id, truth.x, truth.y, truth.w, truth.h);
+            //printf(" mask_n = %d, l.output[obj_index] = %f, l.output[class_index + class_id] = %f \n\n", mask_n, l.output[obj_index], l.output[class_index + class_id]);
 
-                        ++(args->count);
-                        ++(args->class_count);
-                        if (all_ious.iou > .5) recall += 1;
-                        if (all_ious.iou > .75) recall75 += 1;
-                    }
-                }
-            }
+            ++(args->count);
+            ++(args->class_count);
+            if (all_ious.iou > .5) recall += 1;
+            if (all_ious.iou > .75) recall75 += 1;
         }
 
         if (l.iou_thresh < 1.0f) {
@@ -723,6 +746,7 @@ void forward_yolo_layer(const layer l, network_state state)
         yolo_args[b].tot_giou_loss = 0;
         yolo_args[b].count = 0;
         yolo_args[b].class_count = 0;
+        yolo_args[b].mloam_ptr = (mloam *)xcalloc(l.max_boxes * 3 + 60, sizeof(mloam));     // multi_label_one_anchor_choose_maxiou
 
         if (pthread_create(&threads[b], 0, process_batch, &(yolo_args[b]))) error("Thread creation failed", DARKNET_LOC);
    }
@@ -736,6 +760,7 @@ void forward_yolo_layer(const layer l, network_state state)
         tot_giou_loss += yolo_args[b].tot_giou_loss;
         count += yolo_args[b].count;
         class_count += yolo_args[b].class_count;
+        free(yolo_args[b].mloam_ptr);
     }
 
     free(yolo_args);