PaddleSeg v2.4.0
michaelowenliu
released this
20 Jan 01:41
·
24 commits
to release/2.4
since this release
New Features
- Release the upgraded interactive segmentation tool, EISeg 0.4, adding Remote Sensing and Medical annotation.
- Publish the PP-HumanSeg paper for portrait segmentation, including semantic connectivity loss and large-scale video conferencing dataset named PP-HumanSeg14K.
- Release PP-HumanMatting for extremely fine-grained human segmentation.
- Provide 2 tutorials for domain adaptation algorithm PixMatch, and Lane Segmentation.
- Add 9 semantic segmentation models: ESPNetV1, ESPNetV2, HRNet_W48_Contrast, DMNet, ENCNet, PFPNNet, FastFCN, BiSeNetV1, and SegMenter.
- Add 2 loss functions: SECrossEntropyLoss,SemanticConnectivityLoss.
- Add 1 transform method: RandomCenterCrop.
- Add 4 medical segmentation datasets: STARE,DRIVE,CHASE DB1,HRF, and their pretrained models in UNet.
- Provide a comprehensive performance table for all segmentation models in PaddleSeg.
新特性
- 发布交互式分割工具EISeg v0.4版本,支持静态图预测,新增遥感、医疗标注垂类方向。
- 发布人像分割论文PP-HumanSeg,并开源连通性学习(SCL)方法和大规模视频会议数据集PP-HumanSeg14K。
- 开源人像发丝级抠图Matting算法,PP-HumanMatting。
- 新增2个分割案例:无监督领域自适应模型PixMatch,车道线分割。
- 新增9个分割模型:ESPNetV1, ESPNetV2, HRNet_W48_Contrast, DMNet, ENCNet, PFPNNet, FastFCN, BiSeNetV1, 以及Transformer系列模型SegMenter。
- 新增2个损失函数:SECrossEntropyLoss,SemanticConnectivityLoss。
- 新增1个数据增强方法:RandomCenterCrop。
- 新增医疗分割数据集STARE,DRIVE,CHASE DB1,HRF,并提供UNet预训练模型。
- 提供分割模型性能对比图,帮助用户全面了解分割模型性能,方便模型选型。
Bug Fix
- Fix the error of saving image in infer.py #1513
- Fix GINet export bugs #1518
- Remove flip_horizontal when aug_eval is True #1559
- Replace item with numpy in segformer #1579
- Add CocoStuff in init.py #1604
- Fix DICE format and a bug #1630
- Solve the TRT inference problem when dynamic shape is not set #1640
- Fix ENCNet export problem #1677
- Fix weighted cross_entropy bug when 255 in label #1499
- Fix CityscapesSOTA incompatibility in PaddlePaddle 2.2 #1595
New Contributors
- @acdart made their first contribution in #1195
- @txyugood made their first contribution in #1132
- @CuberrChen made their first contribution in #1305
- @neonhuang made their first contribution in #1405
- @mmglove made their first contribution in #1495
- @justld made their first contribution in #1490
- @qianbin1989228 made their first contribution in #1538
- @JiehangXie made their first contribution in #1481
- @Fate-Sunshine made their first contribution in #1549
- @0x45f made their first contribution in #1527
- @ucsk made their first contribution in #1584
- @louis-she made their first contribution in #1604
- @simuler made their first contribution in #1625
- @ETTR123 made their first contribution in #1638
- @JamesLim-sy made their first contribution in #1683
- @Thudjr made their first contribution in #1712
- @Mrwow made their first contribution in #1721
Full Changelog: v2.2.0...v2.4