MaybeShewill-CV / bisenetv2-tensorflow

Unofficial tensorflow implementation of real-time scene image segmentation model "BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation"
https://maybeshewill-cv.github.io/bisenetv2-tensorflow/
MIT License
224 stars 59 forks source link

针对小目标的识别分割应该从哪里下手优化 #52

Closed dgliu123 closed 2 years ago

dgliu123 commented 3 years ago

你好 ,针对小目标的识别分割应该从哪里下手优化