与DEVO相比,我们提出的DEIO将姿态跟踪误差降低了高达71%,这得益于学习和传统优化方法的有机结合。 单目HKU数据集:该数据集是使用DAVIS346(346×260,事件、图像和IMU传感器)以及运动捕捉系统(VICON)收集的,以获得姿态的真实值。为了减轻VICON相机的主动红外(IR)发射器对事件相机产生的显著影响,使用了IR滤镜镜头来消除...
立体HKU数据集:该数据集包含60Hz的立体事件数据和30Hz的立体图像帧,分辨率为346×260,以及1000Hz的IMU数据。它包括手持序列,包括快速运动和HDR场景。在表III中,我们的方法在平均定位误差方面优于所有先前的工作。此外,DEIO击败了DEVO,并将序列的平均准确性提高了多达48%,尤其是在agg walk和agg flip上。DEVO由于仅依...
项目地址:https://github.com/arclab-hku/DEIO 动机(Motivation) 本文的动机源于现有视觉同时定位与建图(SLAM)系统在恶劣条件下(如运动模糊、低光照和高动态范围环境)可靠性和精确性不足的问题。传统基于图像的SLAM方法(如RGB或RGB-D相机)容易受到动态光照变化和运动模糊的影响,而事件相机凭借其高时间分辨率、高动态...
conda activate deio CUDA_VISIBLE_DEVICES=0 PYTHONPATH=${YOUR_WORKSPACE} python script/eval_eio/deio_mono_hku.py \ --datapath=${YOUR_DATAFOLDER} \ --weights=eDBA.pth \ --visual_only=0 \ --evaluate_flag \ --enable_event \ --SCORER_EVAL_USE_GRID \ --side=davis346 ...
arclab-hku/MorAL_Quadruped_RobotsPublic NotificationsYou must be signed in to change notification settings Fork1 Star9 main 1Branch0Tags Code Folders and files Name Last commit message Last commit date Latest commit zerenluo123 small fix Mar 3, 2024 ...
3.1 Run on HKU-dataset 3.1.1 Download our rosbag files (HKU-dataset) Our datasets for evaluation can be download from our One-drive or Baidu-Disk. We have released totally 9 rosbag files for evaluating ESVIO, with the introduction of these datasets can be found on thispage. ...
3)我们首次实现了四旋翼飞行器可以避免多个快速移动小物体从较近距离距离攻击,且传感器只有一个低成本的深度相机。(物体半径<0.035 m, 速度>5 m/s,距离无人机<5m) bilibili 视频链接: b站视频链接 github 链接(即将开源): github.com/arclab-hku/f 能这样躲避子弹的无人机,有没有让你想起曾经风靡一时的游...
SOPVO + object tracking (https://github.com/arclab-hku/ICRA2021tracking): UAV platform: lattepanda + D435i Maintainer: I would like to thank Dr. Shengyang Chen(Dept.ME,PolyU) for his assistant in coding. Cite: R. Duan, D. P. Paudel, C. Fu and P. Lu, "Stereo Orientation Prior ...
arclab-hku/fast-dodgingPublic NotificationsYou must be signed in to change notification settings Fork1 Star27 main 1Branch0Tags Code Latest commit Cannot retrieve latest commit at this time. History 11 Commits sort_ros init Dec 24, 2022
Period: 1 week Overview 0 Active pull requests 0 Active issues 0 Merged pull requests 0 Open pull requests 0 Closed issues 0 New issues There hasn’t been any commit activity on arclab-hku/fast-dodging in the last week. Want to help out? Fork this repository Footer...