Closed nnnetizen closed 4 years ago
@VictorAtPL so does that mean that time distributed layer in this case is only used for handling a 5D tensor as though its a 4D tensor
@nnnetizen
Let's omit batch
axis for simplicity. In case of Mask RCNN (RoIs), instead of applying convolution on whole 4-D tensor [num_rois, POOL_SIZE, POOL_SIZE, channels]
TimeDistributed
applies convolution with same weights for all num_rois
(3-D tensors). In other words, convolution is applied with same weights on tensors of shape [POOL_SIZE, POOL_SIZE, channels]
num_rois
-times.
Further read: https://medium.com/smileinnovation/how-to-work-with-time-distributed-data-in-a-neural-network-b8b39aa4ce00
@VictorAtPL Thanks I think I understand now
Author: Anurag Garg Source: https://intellipaat.com/community/6298/what-is-the-role-of-timedistributed-layer-in-keras?show=6431#a6431
RoIs are for example of shape:
[batch, num_rois, POOL_SIZE, POOL_SIZE, channels]
, so in order to doConv2D
over second dimension (num_rois
), you need to useTimeDistributed
layer, I believe.