from megengine import hub
model = hub.load(
    "megengine/models",
    "retinanet_res50_coco_3x_800size",
    pretrained=True,
    use_cache=False,
)
model.eval()

models_api = hub.import_module(
    "megengine/models",
    git_host="github.com",
)

所有预训练模型希望数据被正确预处理。 模型要求输入BGR的图片, 同时需要等比例缩放到：短边和长边分别不超过800/1333 最后做归一化处理 (均值为: [103.530, 116.280, 123.675], 标准差为: [57.375, 57.120, 58.395])。

下面是一段处理一张图片的样例代码。

# Download an example image from the megengine data website
import urllib
url, filename = ("https://data.megengine.org.cn/images/cat.jpg", "cat.jpg")
try: urllib.URLopener().retrieve(url, filename)
except: urllib.request.urlretrieve(url, filename)

# Read and pre-process the image
import cv2
import megengine as mge

image = cv2.imread("cat.jpg")
data, im_info = models_api.DetEvaluator.process_inputs(image, 800, 1333)
predictions = model(image=mge.tensor(data), im_info=mge.tensor(im_info))
print(predictions)

模型描述

目前我们提供了在COCO2017数据集上预训练的RetinaNet模型, 性能如下：

参考文献

• Focal Loss for Dense Object Detection Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He and Piotr Dollár. IEEE International Conference on Computer Vision (ICCV), 2017.
• Microsoft COCO: Common Objects in Context Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C Lawrence Zitnick. European Conference on Computer Vision (ECCV), 2014.