Figure 2 | Scientific Reports

Figure 2

From: Multiple visual objects are represented differently in the human brain and convolutional neural networks

Figure 2

Single unit response amplitude distribution results. (a) Single voxel response to object pairs plotted against the sum of the individual object responses in LO for all of the participants for both the top 75 most reliable voxels and all the voxels. Units are in beta weights. (b) Responses of the five CNNs (pretrained on the original ImageNet images) to paired and single objects on gray background. (c) Responses of Resnet-50 to paired and single objects on gray background. Resnet-50 was pretrained either with the original ImageNet images (RN50-IN), the stylized ImageNet Images (RN50-SIN), both the original and the stylized ImageNet Images (RN50-SININ), or both sets of images, and then fine-tuned with the stylized ImageNet images (RN50-SININ-IN).

Back to article page