bytetrack 内存泄露问题

背景

bytetrack 在生产运行过程中，发现有内存泄露问题，程序跑了几天，通过free -m观察，内存验证不足
走查完bytetrack源码，发现有内存泄露bug，目前bytetrack官方已经没有维护就不提pr了

c++问题代码

ByteTrack-main\deploy\TensorRT\cpp\src\BYTETracker.cpp
内存泄露在 removed_stracks

  // Step 5: Update state //for (int i = 0; i < this->lost_stracks.size(); i++) {if (this->frame_id - this->lost_stracks[i].end_frame() >this->max_time_lost) {this->lost_stracks[i].mark_removed();removed_stracks.push_back(this->lost_stracks[i]);}}for (int i = 0; i < this->tracked_stracks.size(); i++) {if (this->tracked_stracks[i].state == TrackState::Tracked) {tracked_stracks_swap.push_back(this->tracked_stracks[i]);}}this->tracked_stracks.clear();this->tracked_stracks.assign(tracked_stracks_swap.begin(),tracked_stracks_swap.end());this->tracked_stracks =joint_stracks(this->tracked_stracks, activated_stracks);this->tracked_stracks = joint_stracks(this->tracked_stracks, refind_stracks);this->lost_stracks = sub_stracks(this->lost_stracks, this->tracked_stracks);for (int i = 0; i < lost_stracks.size(); i++) {this->lost_stracks.push_back(lost_stracks[i]);}this->lost_stracks = sub_stracks(this->lost_stracks, this->removed_stracks);for (int i = 0; i < removed_stracks.size(); i++) {this->removed_stracks.push_back(removed_stracks[i]);}

修复c++代码

只保留 max_time_lost *10 范围内的跟踪对象，超过就删除

  // 更新 removed_stracks，只保留在 max_time_lost *10 范围内的跟踪对象std::vector<STrack> filtered_removed_stracks;for (auto& track : this->removed_stracks) {if (this->frame_id - track.end_frame() < 10 * this->max_time_lost) {filtered_removed_stracks.push_back(track);}}this->removed_stracks = filtered_removed_stracks;

python问题代码

yolox/tracker/byte_tracker.py

        """ Step 5: Update state"""for track in self.lost_stracks:if self.frame_id - track.end_frame > self.max_time_lost:track.mark_removed()removed_stracks.append(track)# print('Ramained match {} s'.format(t4-t3))self.tracked_stracks = [t for t in self.tracked_stracks if t.state == TrackState.Tracked]self.tracked_stracks = joint_stracks(self.tracked_stracks, activated_starcks)self.tracked_stracks = joint_stracks(self.tracked_stracks, refind_stracks)self.lost_stracks = sub_stracks(self.lost_stracks, self.tracked_stracks)self.lost_stracks.extend(lost_stracks)self.lost_stracks = sub_stracks(self.lost_stracks, self.removed_stracks)self.removed_stracks.extend(removed_stracks)self.tracked_stracks, self.lost_stracks = remove_duplicate_stracks(self.tracked_stracks, self.lost_stracks)# get scores of lost tracksoutput_stracks = [track for track in self.tracked_stracks if track.is_activated]return output_stracks

修复python代码

在removed_stracks后面添加类似c++的处理方式

        self.removed_stracks.extend(removed_stracks)self.removed_stracks = [track for track in self.removed_stracks ifself.frame_id - track.end_frame < 10 * self.max_time_lost]