目标检测和语义分割计算损失的区别

目标检测，负样本anchor不纳入损失计算。语义分割。背景处像素纳入损失计算。

比如，SSD损失计算相关代码

N = max(1, num_foreground)
        return {
            "bbox_regression": bbox_loss.sum() / N,
            "classification": (cls_loss[foreground_idxs].sum() + cls_loss[background_idxs].sum()) / N
        }

比如，deeplabv3损失计算相关代码

def criterion(inputs, target):
    '''
    cityscapes语义标注，255表示背景处像素
    '''
    target = torch.where(torch.ne(target, 255), target, 19) # 将255转换成19，纳入损失函数计算
    losses = {}
    for name, x in inputs.items():
        losses[name] = nn.functional.cross_entropy(x, target)

    if len(losses) == 1:
        return losses["out"]

    return losses["out"] + 0.5 * losses["aux"]