CCAM
CCAM copied to clipboard
CVI-SZU
[CVPR 2022] C2AM: Contrastive learning of Class-agnostic Activation Map for Weakly Supervised Object Localization and Semantic Segmentation
class SimMaxLoss(nn.Module): def __init__(self, metric='cos', alpha=0.25, reduction='mean'): super(SimMaxLoss, self).__init__() self.metric = metric self.alpha = alpha self.reduction = reduction def forward(self, embedded_bg): """ :param embedded_fg: [N, C] :param embedded_bg: [N, C]...
the backbone of CAM and CCAM is the same or different one? when i use the ccam to refine the CAM, should i start a new network to generate the...
Hello! First of all, you did a great job! Congratulations! Second, I have three questions I'd like to ask you about running the code. 1. First, in the case of...
Do you use other forms of supervision in the process of refining background cues？ And could you release the code for this process?
WSOL
Does the WSOL mentioned in the paper localize one object in the image at a time or can it localize to multiple objects in the image?
Is is true that the backbone will also be retrained from scratch in the WSSS task?
After running the inference, I suppose I will obtain some background images. These images will be used as a cue to do the semantic segmentation. How could I run the...
1.How to determine the boundary area of the object of interest? 2.The paper mentions: the class-agnostic activation maps can be used to extract class-agnostic object bounding boxes for localization. How...