Questions tagged [r-cnn]

For questions related to the family of models known as R-CNN (such as the original R-CNN model, fast R-CNN, faster R-CNN and mask R-CNN).

Filter by
Sorted by
Tagged with
0 votes
0 answers
23 views

In traditional R-CNN, why do we use an SVM after training a ConvNet with a log loss?

I'm learning about development of object detection algorithms and came across this fact which seems strange to me - In the multi-stage pipeline of R-CNN, after using a region extractor to crop the ...
user avatar
1 vote
0 answers
23 views

Why are the learned offsets of anchor boxes in the RCNN object detection models scale invariant?

In the original RCNN paper (https://arxiv.org/pdf/1311.2524.pdf) and continued in later RCNN papers such as faster RCNN (https://arxiv.org/pdf/1506.01497.pdf) the learned offsets of the anchor boxes ...
user avatar
  • 143
1 vote
1 answer
26 views

What to do when the ROIs are smaller than $227 \times 227$ in R-CNN?

As English is not my native language, I have some hard time understanding the following sentence: Regardless of the size or aspect ratio of the candidate region, we warp all pixels in a tight ...
user avatar
2 votes
1 answer
55 views

Is intersection of labels acceptable in computer vision?

I have a dataset, where objects are very close to each other. So, the question is: what is the best approach to label them? There are two possible options: mark objects so that they will not ...
user avatar
2 votes
1 answer
178 views

Is it possible to pre-train a CNN in a self-supervised way so that it can later be used to solve an instance segmentation task?

I would like to use self-supervised learning (SSL) to learn features from images (the dataset consists of similar images with small differences), then use the resulting trained model to bootstrap an ...
user avatar
2 votes
0 answers
33 views

How is the data labelled in order to train a region proposal network?

I don't get how the training of the RPN works. From the forward propagation, I have $W \times H \times k$ outputs from the RPN. How is the training data labeled such that I can use the loss function ...
user avatar
1 vote
0 answers
35 views

Does the selective search algorithm in object detection learn?

I am trying to get a better grasp of how object detection works. I (almost) completely understand the concept behind RPNs. However, I am a little bit confused with the selective search algorithm part. ...
user avatar
2 votes
0 answers
155 views

Inaccurate masks with Mask-RCNN: Stairs effect and sudden stops

I've been using matterport's Mask R-CNN to train on a custom dataset. However, there seem to be some parameters that i failed to correctly define because on practically all of the images, the bottom ...
user avatar
  • 33
2 votes
1 answer
104 views

Why are RNNs used in some computer vision problems?

I am learning computer vision. When I was going through implementations of various computer vision projects, some OCR problems used GRU or LSTM, while some did not. I understand that RNNs are used ...
user avatar
5 votes
1 answer
145 views

How does the region proposal method work in Fast R-CNN?

I read so many articles and the Fast R-CNN paper, but I'm still confused about how the region proposal method works in Fast R-CNN. As you can see in the image below, they say they used a proposal ...
user avatar
  • 51
1 vote
1 answer
163 views

In Fast R-CNN, how are input RoIs mapped to the respective RoIs in the feature map before RoI pooling?

I've been reading the Fast R-CNN paper. My understanding is that the input to one forward pass is the whole input image plus a list of RoIs (generated by selective search or another region proposal ...
user avatar
1 vote
1 answer
578 views

In Faster R-CNN, how can I get the predicted bounding box given the neural network's output?

The RPN loss in Faster RCNN paper is $$ L({p_i}, {t_i}) = \frac{1}{N_{cls}} \sum_{i} L_{cls}(p_i,p_i^*) + \lambda \frac{1}{N_{reg}} \sum_i p_i^* L_{reg}(t_i, t_i^*) $$ For regression problems, we have ...
user avatar