78 Quantifying and Learning Linear Symmetry-Based Disentanglement (LSBD) Table 4.1: Encoder and decoder architectures used in most methods. ENCODER INPUT SIZE (64,64, NUMBER CHANNELS) CONV FILTERS 32, KERNEL 4, STRIDE 2, RELU CONV FILTERS 32, KERNEL 4, STRIDE 2, RELU CONV FILTERS 64, KERNEL 4, STRIDE 2, RELU CONV FILTERS 64, KERNEL 4, STRIDE 2, RELU DENSE UNITS 256, RELU DENSE(X2) UNITS DEPEND ON METHOD DECODER INPUT SIZE (NUMBER OF LATENT DIMENSIONS) DENSE UNITS 256, RELU DENSE UNITS 4*4*64, RELU RESHAPE (4,4,64) CONVT FILTERS 64, KERNEL 4, STRIDE 2, RELU CONVT FILTERS 32, KERNEL 4, STRIDE 2, RELU CONVT FILTERS 32, KERNEL 4, STRIDE 2, RELU CONVT FILTERS (NUMBER CHANNELS), KERNEL 4, STRIDE 2, SIGMOID Table 4.2: Encoder and decoder architecture used to train LSBD-VAE/0 for ModelNet40 dataset. ENCODER INPUT SIZE (64, 64, NUMBER CHANNELS) DENSE UNITS 512, RELU, BATCH NORMALIZATION DENSE UNITS 256, RELU, BATCH NORMALIZATION DENSE(X2) UNITS DEPEND ON METHOD DECODER INPUT SIZE (NUMBER OF LATENT DIMENSIONS) DENSE UNITS 256, RELU, BATCH NORMALIZATION DENSE UNITS 512, RELU, BATCH NORMALIZATION DENSE UNITS 64*64*NUMBER OF CHANNELS, SIGMOID RESHAPE (64, 64, NUMBER OF CHANNELS)
RkJQdWJsaXNoZXIy MjY0ODMw