Improved tracker parametrization.

rpt/tracker.hpp

@@ -24,7 +24,7 @@ struct Track
 class PoseTracker
 {
 public:
-    PoseTracker(float fps);
+    PoseTracker(float max_movement_speed, float max_distance);
 
     std::vector<std::tuple<size_t, std::vector<std::array<float, 4>>>> track_poses(
         const std::vector<std::vector<std::array<float, 4>>> &poses_3d,
@@ -35,15 +35,9 @@ public:
 
 private:
     float max_distance;
+    float max_movement_speed;
     size_t history_size = 3;
 
-    // Approach speed of EN ISO 13855 with 2000 mm/sec for hand speed
-    // and an additional factor to compensate for noise-based jumps
-    float max_movement_speed = 2.0 * 1.5;
-
-    // The size of an A4 sheet of paper which is assumed to fit between two different persons
-    float max_base_distance = 0.3;
-
     std::vector<double> timestamps;
     std::vector<Track> pose_tracks;
 
@@ -70,9 +64,10 @@ private:
 // =================================================================================================
 // =================================================================================================
 
-PoseTracker::PoseTracker(float fps)
+PoseTracker::PoseTracker(float max_movement_speed, float max_distance)
 {
-    this->max_distance = max_base_distance + max_movement_speed / fps;
+    this->max_movement_speed = max_movement_speed;
+    this->max_distance = max_distance;
 }
 
 // =================================================================================================

scripts/test_skelda_dataset.cpp

@@ -123,6 +123,8 @@ int main(int argc, char **argv)
     const size_t min_group_size = config["min_group_size"];
     const int take_interval = config["take_interval"];
     const float ifps = config["fps"];
+    const float max_movement_speed = config["max_movement_speed"];
+    const float max_track_distance = config["max_track_distance"];
 
     // Load 2D model
     bool use_wb = utils_pipeline::use_whole_body(whole_body);
@@ -134,7 +136,7 @@ int main(int argc, char **argv)
     std::unique_ptr<Triangulator> tri_model = std::make_unique<Triangulator>(
         min_match_score, min_group_size);
     std::unique_ptr<PoseTracker> pose_tracker = std::make_unique<PoseTracker>(
-        ifps);
+        max_movement_speed, max_track_distance);
 
     // Timers
     size_t time_count = dataset.size();
@@ -216,7 +218,7 @@ int main(int argc, char **argv)
         auto &item = dataset[i];
         auto &poses_2d = all_poses_2d[i];
 
-        if (old_scene != item["scene"] || old_id + take_interval < item["index"])
+        if (old_scene != item["scene"] || old_id + take_interval != item["index"])
         {
             // Reset last poses if scene changes
             tri_model->reset();
@@ -248,7 +250,7 @@ int main(int argc, char **argv)
         elapsed = std::chrono::high_resolution_clock::now() - stime;
         times_pose3d.push_back(elapsed.count());
 
-        if (ifps < 1)
+        if (ifps <= 0)
         {
             // Disable pose tracking if frame rate is too low
             times_tracks.push_back(0.0);

scripts/test_skelda_dataset.py

@@ -51,6 +51,15 @@ default_min_group_size = 1
 # If most of the time only one person is in a image, disable it, because it is slightly slower then
 default_batch_poses = True
 
+# Approach speed of EN ISO 13855 with 2000 mm/sec for hand speed
+# and an additional factor to compensate for noise-based jumps
+default_max_movement_speed = 2.0 * 1.5
+
+# The size of an A4 sheet of paper which is assumed to fit between two different persons
+# and additionally the distance a person can move between two frames (here at 10 fps)
+default_max_track_distance = 0.3 + default_max_movement_speed / 10
+
+
 datasets = {
     "human36m": {
         "path": "/datasets/human36m/skelda/pose_test.json",
@@ -61,6 +70,8 @@ datasets = {
         "min_bbox_score": 0.4,
         "min_bbox_area": 0.1 * 0.1,
         "batch_poses": False,
+        "max_movement_speed": 2.0 * 1.5,
+        "max_track_distance": 0.3 + default_max_movement_speed / (50 / 5),
     },
     "panoptic": {
         "path": "/datasets/panoptic/skelda/test.json",
@@ -69,16 +80,19 @@ datasets = {
         # "cams": ["00_03", "00_06", "00_12", "00_13", "00_23", "00_15", "00_10", "00_21", "00_09", "00_01"],
         # "cams": [],
         "take_interval": 3,
+        "fps": 30,
         "min_match_score": 0.95,
         "use_scenes": ["160906_pizza1", "160422_haggling1", "160906_ian5"],
         "min_group_size": 1,
         # "min_group_size": 4,
         # "min_group_size": 11,
         "min_bbox_area": 0.05 * 0.05,
+        "max_track_distance": 0.3 + default_max_movement_speed / (30 / 3),
     },
     "mvor": {
         "path": "/datasets/mvor/skelda/all.json",
         "take_interval": 1,
+        "fps": -1,
         "with_depth": False,
         "min_match_score": 0.85,
         "min_bbox_score": 0.25,
@@ -89,6 +103,7 @@ datasets = {
         "take_interval": 1,
         "min_match_score": 0.92,
         "min_bbox_score": 0.5,
+        "max_track_distance": 0.3 + default_max_movement_speed / 25,
     },
     "shelf": {
         "path": "/datasets/shelf/skelda/test.json",
@@ -96,82 +111,108 @@ datasets = {
         "fps": 25,
         "min_match_score": 0.95,
         "min_group_size": 2,
+        "max_track_distance": 0.3 + default_max_movement_speed / 25,
     },
     "ikeaasm": {
         "path": "/datasets/ikeaasm/skelda/test.json",
         "take_interval": 2,
+        "fps": -1,
         "min_match_score": 0.92,
         "min_bbox_score": 0.20,
     },
     "chi3d": {
         "path": "/datasets/chi3d/skelda/all.json",
         "take_interval": 5,
+        "fps": 50,
+        "max_track_distance": 0.3 + default_max_movement_speed / (50 / 5),
     },
     "tsinghua": {
         "path": "/datasets/tsinghua/skelda/test.json",
         "take_interval": 3,
+        "fps": 30,
         "min_match_score": 0.95,
         "min_group_size": 2,
+        "max_track_distance": 0.3 + default_max_movement_speed / (30 / 3),
     },
     "human36m_wb": {
         "path": "/datasets/human36m/skelda/wb/test.json",
         "take_interval": 100,
+        "fps": -1,
         "min_bbox_score": 0.4,
         "batch_poses": False,
     },
     "egohumans_tagging": {
         "path": "/datasets/egohumans/skelda/all.json",
         "take_interval": 2,
+        "fps": 20,
         "subset": "tagging",
         "min_match_score": 0.92,
         "min_group_size": 2,
         "min_bbox_score": 0.2,
         "min_bbox_area": 0.05 * 0.05,
+        "max_movement_speed": 4.0 * 1.5,
+        "max_track_distance": 0.3 + (4.0 * 1.5) / (20 / 2),
     },
     "egohumans_legoassemble": {
         "path": "/datasets/egohumans/skelda/all.json",
         "take_interval": 2,
+        "fps": 20,
         "subset": "legoassemble",
         "min_group_size": 2,
+        "max_track_distance": 0.3 + default_max_movement_speed / (20 / 2),
     },
     "egohumans_fencing": {
         "path": "/datasets/egohumans/skelda/all.json",
         "take_interval": 2,
+        "fps": 20,
         "subset": "fencing",
         "min_group_size": 7,
         "min_bbox_score": 0.5,
         "min_bbox_area": 0.05 * 0.05,
+        "max_track_distance": 0.3 + default_max_movement_speed / (20 / 2),
     },
     "egohumans_basketball": {
         "path": "/datasets/egohumans/skelda/all.json",
         "take_interval": 2,
+        "fps": 20,
         "subset": "basketball",
         "min_group_size": 7,
         "min_bbox_score": 0.25,
         "min_bbox_area": 0.025 * 0.025,
+        "max_movement_speed": 4.0 * 1.5,
+        "max_track_distance": 0.3 + (4.0 * 1.5) / (20 / 2),
     },
     "egohumans_volleyball": {
         "path": "/datasets/egohumans/skelda/all.json",
         "take_interval": 2,
+        "fps": 20,
         "subset": "volleyball",
         "min_group_size": 11,
         "min_bbox_score": 0.25,
         "min_bbox_area": 0.05 * 0.05,
+        "max_movement_speed": 4.0 * 1.5,
+        "max_track_distance": 0.3 + (4.0 * 1.5) / (20 / 2),
     },
     "egohumans_badminton": {
         "path": "/datasets/egohumans/skelda/all.json",
         "take_interval": 2,
+        "fps": 20,
         "subset": "badminton",
         "min_group_size": 7,
         "min_bbox_score": 0.25,
         "min_bbox_area": 0.05 * 0.05,
+        "max_movement_speed": 4.0 * 1.5,
+        "max_track_distance": 0.3 + (4.0 * 1.5) / (20 / 2),
     },
     "egohumans_tennis": {
         "path": "/datasets/egohumans/skelda/all.json",
         "take_interval": 2,
+        "fps": 20,
         "subset": "tennis",
         "min_group_size": 11,
         "min_bbox_area": 0.025 * 0.025,
+        "max_movement_speed": 4.0 * 1.5,
+        "max_track_distance": 0.3 + (4.0 * 1.5) / (20 / 2),
     },
 }
 
@@ -338,6 +379,12 @@ def main():
     min_bbox_score = datasets[dataset_use].get("min_bbox_score", default_min_bbox_score)
     min_bbox_area = datasets[dataset_use].get("min_bbox_area", default_min_bbox_area)
     batch_poses = datasets[dataset_use].get("batch_poses", default_batch_poses)
+    max_movement_speed = datasets[dataset_use].get(
+        "max_movement_speed", default_max_movement_speed
+    )
+    max_track_distance = datasets[dataset_use].get(
+        "max_track_distance", default_max_track_distance
+    )
 
     # Save config
     config_path = tmp_export_dir + "config.json"
@@ -347,6 +394,8 @@ def main():
         "min_bbox_score": min_bbox_score,
         "min_bbox_area": min_bbox_area,
         "batch_poses": batch_poses,
+        "max_movement_speed": max_movement_speed,
+        "max_track_distance": max_track_distance,
         "whole_body": whole_body,
         "take_interval": datasets[dataset_use]["take_interval"],
         "fps": datasets[dataset_use]["fps"],