본문 바로가기

Study/event-based camera

HASTE 설치

https://github.com/ialzugaray/haste

Dependencies

선행 ) GLOG 설치

git clone https://github.com/google/glog.git
cd glog

# conda deactivate
cmake -S . -B build -G "Unix Makefiles"
cmake --build build
# cmake --build build --target test

에러 해결법 → conda deactivate

콘다를 설치하고 그냥 실행하면 에러가 생긴다.

https://github.com/google/glog/issues/843
정상설치 확인

Install only stand-alone benchmarking app

git clone https://github.com/ialzugaray/haste.git
git submodule update --init --recursive
mkdir -p haste/build && cd haste/build
cmake .. -DCMAKE_BUILD_TYPE=Release -DGUI=false
make

빌드에서 자꾸 glog가 문제 생긴다!!!!! 으아아 열받아

https://github.com/google/glog/issues/519

이게 좀 도움될 것 같은데 안돼,,,, 설치경로를 인식못하는 것 같아!!!

해결!

# 원본 
find_package(Glog REQUIRED)
include_directories(${GLOG_INCLUDE_DIRS})

#수정&해결!
find_package(PkgConfig REQUIRED)
pkg_check_modules(glog REQUIRED libglog)
#find_package(Glog REQUIRED)
include_directories(${GLOG_INCLUDE_DIRS})

install only stand-alone benchmarking & GUI apps

git clone https://github.com/ialzugaray/haste.git
git submodule update --init --recursive
mkdir -p haste/build && cd haste/build

sudo apt-get install mesa-utils libglfw3-dev

cmake .. -DCMAKE_BUILD_TYPE=Release -DGUI=true
make

앞선 glog 문제와 동일. 해결 완료.

GUI app

./tracking_app_gui --events_file=path/to/dataset/events.txt --camera_params_file=path/to/dataset/calib.txt

./tracking_app_gui --events_file=/home/burger/eventVO/haste_example/events.txt --camera_params_file=/home/burger/eventVO/haste_example/calib.txt

Stand-alone benchmarking app

./tracking_app_file --events_file=path/to/dataset/events.txt        `# Load events from specified file.`\
                    --camera_params_file=path/to/dataset/calib.txt  `# Apply undistortion as pre-processing step.`\
                    --seed=0.6,125.0,52.0,0.0,0                     `# Seed tracker #0 at 0.6s, in (125.0,52.0) pixel coordiates with initial orientation 0 degrees.`'\
                    --centered_initialization=false                 `# Do not enforce centered initialization. Track will start arbitrarily later than seed.`'\
                    --tracker_type=haste_correlation_star           `# Employed tracker is HasteCorrelation*.`\
                    --visualize=false                               `# Show intermediate tracking states as images.`

./tracking_app_file --events_file=/home/burger/eventVO/haste_example/events.txt       `# Load events from specified file.`\
                    --camera_params_file=/home/burger/eventVO/haste_example/calib.txt  `# Apply undistortion as pre-processing step.`\
                    --seed=0.6,125.0,52.0,0.0,0                     `# Seed tracker #0 at 0.6s, in (125.0,52.0) pixel coordiates with initial orientation 0 degrees.`'\
                    --centered_initialization=false                 `# Do not enforce centered initialization. Track will start arbitrarily later than seed.`'\
                    --tracker_type=haste_correlation_star           `# Employed tracker is HasteCorrelation*.`\
                    --visualize=false                               `# Show intermediate tracking states as images.`


Uploaded by N2T

'Study > event-based camera' 카테고리의 다른 글

e2calib 설치 및 사용  (2) 2023.03.06
E2VID 설치 및 사용  (0) 2023.01.12