@@ -12,7 +12,11 @@ PaddleDetection supported practical projects such as industrial quality inspecti
...
@@ -12,7 +12,11 @@ PaddleDetection supported practical projects such as industrial quality inspecti
image object detection, and automatic inspection with its practical features such as model compression
image object detection, and automatic inspection with its practical features such as model compression
and multi-platform deployment.
and multi-platform deployment.
**Now all models in PaddleDetection require PaddlePaddle version 1.7 or higher, or suitable develop version.**
[PP-YOLO](https://arxiv.org/abs/2007.12099), which is faster and has higer performance than YOLOv4,
has been released, it reached mAP(0.5:0.95) as 45.2% on COCO test2019 dataset and 72.9 FPS on single
Test V100. Please refer to [PPYOLO](configs/ppyolo/README.md) for details.
**Now all models in PaddleDetection require PaddlePaddle version 1.8 or higher, or suitable develop version.**
<divalign="center">
<divalign="center">
<imgsrc="docs/images/000000570688.jpg"/>
<imgsrc="docs/images/000000570688.jpg"/>
...
@@ -75,6 +79,7 @@ More models:
...
@@ -75,6 +79,7 @@ More models:
- FCOS
- FCOS
- CornerNet-Squeeze
- CornerNet-Squeeze
- YOLOv4
- YOLOv4
- PP-YOLO
More Backbones:
More Backbones:
...
@@ -163,7 +168,7 @@ The following is the relationship between COCO mAP and FPS on Tesla V100 of SOTA
...
@@ -163,7 +168,7 @@ The following is the relationship between COCO mAP and FPS on Tesla V100 of SOTA
PaddleDetection is released under the [Apache 2.0 license](LICENSE).
PaddleDetection is released under the [Apache 2.0 license](LICENSE).
## Updates
## Updates
v0.3.0 was released at `05/2020`, add anchor-free, EfficientDet, YOLOv4, etc. Launched mobile and server-side practical and efficient multiple models. For example, the YOLOv3-MobileNetv3 mobile side model is accelerated 3.5 times, the server side has optimized the two-stage model, and the speed and accuracy have high cost performance. We also refactored predictive deployment functions, and improved ease of use, fix many known bugs, etc.
v0.4.0 was released at `05/2020`, add PP-YOLO, TTFNet, HTC, ACFPN, etc. And add BlaceFace face landmark detection model, add a series of optimized SSDLite models on mobile side, add data augmentations GridMask and RandomErasing, add Matrix NMS and EMA training, and improved ease of use, fix many known bugs, etc.
Please refer to [版本更新文档](docs/CHANGELOG.md) for details.
Please refer to [版本更新文档](docs/CHANGELOG.md) for details.