Closed hxp33 closed 5 years ago
sure
can we use yolo for pedestrians and vehicle detection in low light or night time, if we have thermal image dataset. Thermal image dataset link https://www.flir.com/oem/adas/adas-dataset-form/
Anyone has done it? How is the performance?
Where I can find a good tutorial or how to train with this dataset ? I want to transfer learning from the already trained dataset. And train person and cars categories from thermal.
@88madri , did You find a tutorial? Thanks!
We are working on it. However we have designed our own architecture for thermal images. Still testing our model with different set of thermal images.
I am also working on it
How well have You seen it generalise on new data?
Martin
K, 20. märts 2019 09.03 Bibin wilson notifications@github.com kirjutas:
I am also working on it
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/pjreddie/darknet/issues/261#issuecomment-474711236, or mute the thread https://github.com/notifications/unsubscribe-auth/ALnA1yYdoA1CTVoh_BwY49OnqpZZcylVks5vYd06gaJpZM4QBZEL .
I want to detect person from thermal infrared images , can yolo work on it. The dataset i used . Or can you give any suggestion