S

Segformer B5 Finetuned 100by100PNG 50epochs Attempt2 100epochs Backgroundclass 2

Developed by JCAI2000
This model is an image segmentation model fine-tuned on the JCAI2000/100By100BranchPNG dataset based on nvidia/mit-b5, primarily used for semantic segmentation tasks of branches and backgrounds.
Downloads 16
Release Time : 9/6/2023

Model Overview

This is a semantic segmentation model based on the SegFormer architecture, fine-tuned specifically for processing 100x100 pixel PNG images, capable of accurately distinguishing between branch and background regions in images.

Model Features

High-precision segmentation
Achieved a mean Intersection over Union (mIoU) of 0.8933 and an average accuracy of 0.9531 on the evaluation set
Strong background recognition
Background recognition accuracy reached 0.9732 with an IoU of 0.9597
Optimized for small-size images
Specifically optimized for small 100x100 pixel PNG images

Model Capabilities

Image semantic segmentation
Background recognition
Branch detection

Use Cases

Agricultural image analysis
Plant branch detection
Used to identify and analyze branch structures in plant images
Branch recognition accuracy 93.3%, IoU 82.7%
Remote sensing image processing
Land cover classification
Used to distinguish land cover types such as vegetation and non-vegetation areas
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase