search

Jaraxxus-Me / AirDet

Full conference version of AirDet: Few-Shot Detection without Fine-tuning for Autonomous Exploration
https://jaraxxus-me.github.io/ECCV2022_AirDet/
BSD 2-Clause "Simplified" License
70 stars 10 forks source link

How to train on self-define dataset #9

Closed sixone-Jiang closed 1 year ago

sixone-Jiang commented 1 year ago

At the same time, I have a question about few-shot-object-detection task, if it is based on fine-tuning, can the pre-training model include base-classes? For example, fine-tune by VOC(1-5shot) and Val on VOC(20 classes), but pretrained on Coco 2017(80 classes above voc 20classes). I am looking forward to your reply.

sixone-Jiang commented 1 year ago

And, how can I only inference an image?