3

我正在尝试使用detectron2框架提取类别检测高于某个阈值的区域特征。稍后我将在我的管道中使用这些功能(类似于:VilBert第 3.1 节训练 ViLBERT)到目前为止,我已经使用此配置训练了一个 Mask R-CNN,并在一些自定义数据上对其进行了微调。它表现良好。我想做的是从我训练的模型中提取特征,用于生成的边界框。

编辑:我查看了关闭我帖子的用户所写的内容并试图对其进行改进。尽管读者需要关于我在做什么的上下文。如果您对我如何使问题变得更好有任何想法,或者您对如何做我想做的事情有一些见解,欢迎您提供反馈!

我有个问题:

  1. 为什么我只得到一个预测实例,但是当我查看预测 CLS 分数时,超过 1 个通过阈值?

我相信这是产生 ROI 特征的正确方法:

images = ImageList.from_tensors(lst[:1], size_divisibility=32).to("cuda")  # preprocessed input tensor
#setup config
cfg = get_cfg()
cfg.merge_from_file(model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x.yaml"))
cfg.MODEL.WEIGHTS = os.path.join(cfg.OUTPUT_DIR, "model_final.pth")
cfg.SOLVER.IMS_PER_BATCH = 1
cfg.MODEL.ROI_HEADS.NUM_CLASSES = 1  # only has one class (pnumonia)
#Just run these lines if you have the trained model im memory
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7   # set the testing threshold for this model
#build model
model = build_model(cfg)
DetectionCheckpointer(model).load("output/model_final.pth")
model.eval()#make sure its in eval mode

#run model
with torch.no_grad():
    features = model.backbone(images.tensor.float())
    proposals, _ = model.proposal_generator(images, features)
    instances = model.roi_heads._forward_box(features, proposals)

然后

pred_boxes = [x.pred_boxes for x in instances]
rois = model.roi_heads.box_pooler([features[f] for f in model.roi_heads.in_features], pred_boxes)

这应该是我的 ROI 功能。

我非常困惑的是,除了使用推理时产生的边界框外,我还可以使用提案和提案框及其类分数来获得该图像的前 n 个特征。很酷,所以我尝试了以下方法:

proposal_boxes = [x.proposal_boxes for x in proposals]
proposal_rois = model.roi_heads.box_pooler([features[f] for f in model.roi_heads.in_features], proposal_boxes)
#found here: https://detectron2.readthedocs.io/_modules/detectron2/modeling/roi_heads/roi_heads.html
box_features = model.roi_heads.box_head(proposal_rois)
predictions = model.roi_heads.box_predictor(box_features)
pred_instances, losses = model.roi_heads.box_predictor.inference(predictions, proposals)

我应该在我的预测对象中获得我的建议框功能及其cls。检查这个预测对象,我看到每个框的分数:

预测对象中的 CLS 分数

(tensor([[ 0.6308, -0.4926],
         [-1.6662,  1.5430],
         [-0.2080,  0.4856],
         ...,
         [-6.9698,  6.6695],
         [-5.6361,  5.4046],
         [-4.4918,  4.3899]], device='cuda:0', grad_fn=<AddmmBackward>),

在 softmaxing 并将这些 cls 分数放入数据框中并将阈值设置为 0.6 后,我得到:

pred_df = pd.DataFrame(predictions[0].softmax(-1).tolist())
pred_df[pred_df[0] > 0.6]
    0           1
0   0.754618    0.245382
6   0.686816    0.313184
38  0.722627    0.277373

在我的预测对象中,我得到了相同的最高分,但只有 1 个实例而不是 2 个(我设置了cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7):

预测实例

[Instances(num_instances=1, image_height=800, image_width=800, fields=[pred_boxes: Boxes(tensor([[548.5992, 341.7193, 756.9728, 438.0507]], device='cuda:0',
        grad_fn=<IndexBackward>)), scores: tensor([0.7546], device='cuda:0', grad_fn=<IndexBackward>), pred_classes: tensor([0], device='cuda:0')])]

预测还包含张量:Nx4 或 Nx(Kx4) 边界框回归增量。我不完全知道他们做什么和看起来像:

预测对象中的边界框回归增量

tensor([[ 0.2502,  0.2461, -0.4559, -0.3304],
        [-0.1359, -0.1563, -0.2821,  0.0557],
        [ 0.7802,  0.5719, -1.0790, -1.3001],
        ...,
        [-0.8594,  0.0632,  0.2024, -0.6000],
        [-0.2020, -3.3195,  0.6745,  0.5456],
        [-0.5542,  1.1727,  1.9679, -2.3912]], device='cuda:0',
       grad_fn=<AddmmBackward>)

另一个奇怪的是,我的提案框我的预测框不同但相似:

提案边界框

[Boxes(tensor([[532.9427, 335.8969, 761.2068, 438.8086],#this box vs the instance box
         [102.7041, 352.5067, 329.4510, 440.7240],
         [499.2719, 317.9529, 764.1958, 448.1386],
         ...,
         [ 25.2890, 379.3329,  28.6030, 429.9694],
         [127.1215, 392.6055, 328.6081, 489.0793],
         [164.5633, 275.6021, 295.0134, 462.7395]], device='cuda:0'))]
4

1 回答 1

7

你快到了。查看roi_heads.box_predictor.inference()你会发现它并不仅仅是对候选框的分数进行排序。首先,它应用框增量来重新调整提案框。然后它计算非最大抑制以删除非重叠框(同时还应用其他超设置,例如分数阈值)。最后,它根据得分对前 k 个框进行排名。这可能解释了为什么您的方法产生相同的框分数但输出框的数量及其坐标不同。

回到您最初的问题,这是在一次推理过程中提取提议框的特征的方法:

image = cv2.imread('my_image.jpg')
height, width = image.shape[:2]
image = torch.as_tensor(image.astype("float32").transpose(2, 0, 1))
inputs = [{"image": image, "height": height, "width": width}]
with torch.no_grad():
    images = model.preprocess_image(inputs)  # don't forget to preprocess
    features = model.backbone(images.tensor)  # set of cnn features
    proposals, _ = model.proposal_generator(images, features, None)  # RPN

    features_ = [features[f] for f in model.roi_heads.box_in_features]
    box_features = model.roi_heads.box_pooler(features_, [x.proposal_boxes for x in proposals])
    box_features = model.roi_heads.box_head(box_features)  # features of all 1k candidates
    predictions = model.roi_heads.box_predictor(box_features)
    pred_instances, pred_inds = model.roi_heads.box_predictor.inference(predictions, proposals)
    pred_instances = model.roi_heads.forward_with_given_boxes(features, pred_instances)

    # output boxes, masks, scores, etc
    pred_instances = model._postprocess(pred_instances, inputs, images.image_sizes)  # scale box to orig size
    # features of the proposed boxes
    feats = box_features[pred_inds]
于 2020-07-04T00:15:18.597 回答