search

ViTAE-Transformer / DeepSolo

The official repo for [CVPR'23] "DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting" & [ArXiv'23] "DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spotting"
Other
250 stars 34 forks source link

visualization code for spotting results #11

Closed SleepEarlyLiveLong closed 1 year ago

SleepEarlyLiveLong commented 1 year ago

Could you please release code of the visualization code of spotting results as you showed in Figure 4 and Figure 11? Thank you

ymy-k commented 1 year ago

Figure 4: You can return and save the sampling_locations & attention_weights in adet/layers/ms_deform_attn.py, then plot them on the image. Figure 11: Do you mean the points on center curve? You can find the code here

SleepEarlyLiveLong commented 1 year ago

Figure 4: You can return and save the sampling_locations & attention_weights in adet/layers/ms_deform_attn.py, then plot them on the image. Figure 11: Do you mean the points on center curve? You can find the code here

thanks a lot, I got it.