Does PIPs process each of the S=8 frames independently, without sharing any information between frames?
ZHAOZHIHAO opened this issue · comments
Hi,
I read the paper and code. It seems that PIPs processes the S=8 frames independently, without sharing any information between these frames?
As in Figure 2, the three steps "Initialize positions and appearance features", "Measure local similarity", and "Update positions and features" seem never blend information between different frames.
Best,
The key you might be missing is the MLP-Mixer. Half of its parameters apply within-frame operations, and the other half apply cross-frame operations.
Thanks for the clarification!