Perceptography unveils the causal contribution of inferior temporal cortex to visual perception.

Neurons in the inferotemporal (IT) cortex respond selectively to complex visual features, implying their role in object perception. However, perception is subjective and cannot be read out from neural responses; thus, bridging the causal gap between neural activity and perception demands independent characterization of perception. Historically, though, the complexity of the perceptual alterations induced by artificial stimulation of IT cortex has rendered them impossible to quantify. To address this old problem, we tasked male macaque monkeys to detect and report optical impulses delivered to their IT cortex. Combining machine learning with high-throughput behavioral optogenetics, we generated complex and highly specific images that were hard for the animal to distinguish from the state of being cortically stimulated. These images, named "perceptograms" for the first time, reveal and depict the contents of the complex hallucinatory percepts induced by local neural perturbation in IT cortex. Furthermore, we found that the nature and magnitude of these hallucinations highly depend on concurrent visual input, stimulation location, and intensity. Objective characterization of stimulation-induced perceptual events opens the door to developing a mechanistic theory of visual perception. Further, it enables us to make better visual prosthetic devices and gain a greater understanding of visual hallucinations in mental disorders.

MaskFaceGAN: High-Resolution Face Editing With Masked GAN Latent Code Optimization.

Face editing represents a popular research topic within the computer vision and image processing communities. While significant progress has been made recently in this area, existing solutions: (i) are still largely focused on low-resolution images, (ii) often generate editing results with visual artefacts, or (iii) lack fine-grained control over the editing procedure and alter multiple (entangled) attributes simultaneously, when trying to generate the desired facial semantics. In this paper, we aim to address these issues through a novel editing approach, called MaskFaceGAN that focuses on local attribute editing. The proposed approach is based on an optimization procedure that directly optimizes the latent code of a pre-trained (state-of-the-art) Generative Adversarial Network (i.e., StyleGAN2) with respect to several constraints that ensure: (i) preservation of relevant image content, (ii) generation of the targeted facial attributes, and (iii) spatially-selective treatment of local image regions. The constraints are enforced with the help of an (differentiable) attribute classifier and face parser that provide the necessary reference information for the optimization procedure. MaskFaceGAN is evaluated in extensive experiments on the FRGC, SiblingsDB-HQf, and XM2VTS datasets and in comparison with several state-of-the-art techniques from the literature. Our experimental results show that the proposed approach is able to edit face images with respect to several local facial attributes with unprecedented image quality and at high-resolutions ( 1024×1024 ), while exhibiting considerably less problems with attribute entanglement than competing solutions. The source code is publicly available from: https://github.com/MartinPernus/MaskFaceGAN.