search

fcdl94 / CoMFormer

Official implementation of "CoMFormer: Continual Learning in Semantic and Panoptic Segmentation"
https://arxiv.org/abs/2211.13999
Other
35 stars 2 forks source link

Visual Results on VOC #6

Open XinyueZhangqdu opened 1 month ago

XinyueZhangqdu commented 1 month ago

I found both the experimental results and the innovative aspects deeply intriguing. Could you please provide me with visual results from the testing on the VOC dataset for the configurations 10-1, 2-2, 19-1, 15-5, and 15-1? We would like to reference your paper in our research and discuss the relevant experiments and algorithms. Thank you very much for taking the time to respond.

fcdl94 commented 1 month ago

We actually did not reported experiments on Pascal-VOC but feel free to use the code to obtain them.

XinyueZhangqdu commented 1 month ago

Thank you for your prompt reply. Could you kindly share visual result images from the ADE testing dataset? Unfortunately, I lack the necessary GPU resources to quickly replicate your code. Therefore, I would greatly appreciate it if you could provide visualizations of the results on the ADE dataset.

fcdl94 commented 1 month ago

What do you mean by visual result? Qualitative as in the paper?

XinyueZhangqdu commented 1 month ago

Thank you for your prompt reply. The "visual result" refers to the qualitative results that resemble Figure 4 in the paper. Could you kindly share the segmentation results directly from the test set?

fcdl94 commented 1 month ago

I don't have the results on the test set. I used a script to generate some randomly picked examples. Unfortunately I lost the script but you can write your own following the M2F demo code.

It should be not compute intensive since you only need to use a single image per forward.

XinyueZhangqdu commented 1 month ago

Thank you for your assistance and for answering my questions.