One of the main challenges of social interaction in virtual reality settings is that head-mounted displays occlude a large portion of the face, blocking facial expressions and thereby restricting social engagement cues among users. Hence, auxiliary means of sensing and conveying these expressions are needed. We present an algorithm to automatically infer expressions by analyzing only a partially occluded face while the user is engaged in a virtual reality experience. Specifically, we show that images of the user's eyes captured from an IR gaze-tracking camera within a VR headset are sufficient to infer a select subset of facial expressions without the use of any fixed external camera. Using these inferences, we can generate dynamic avatars in real-time which function as an expressive surrogate for the user. We propose a novel data collection pipeline as well as a novel approach for increasing CNN accuracy via personalization. Our results show a mean accuracy of 74% (F 1 of 0.73) among 5 'emotive' expressions and a mean accuracy of 70% (F 1 of 0.68) among 10 distinct facial action units.
Why do interactions become more hostile when social relations shift from “me versus you” to “us versus them”? One possibility is that acting with a group can reduce spontaneous self-referential processing in the moral domain and, in turn, facilitate competitor harm. We tested this hypothesis in an fMRI experiment in which (i) participants performed a competitive task once alone and once with a group; (ii) spontaneous self-referential processing during competition was indexed unobtrusively by activation in an independently localized region of the medial prefrontal cortex (mPFC) associated with self-reference; and (iii) we assessed participants’ willingness to harm competitors versus teammates. As predicted, participants who showed reduced mPFC activation in response to descriptions of their own moral behaviors while competing in a group were more willing to harm competitors. These results suggest that intergroup competition (above and beyond inter-personal competition) can reduce self-referential processing of moral information, enabling harmful behaviors towards members of a competitive group.
In this work we propose a model that can manipulate individual visual attributes of objects in a real scene using examples of how respective attribute manipulations affect the output of a simulation. As an example, we train our model to manipulate the expression of a human face using nonphotorealistic 3D renders of a face with varied expression. Our model manages to preserve all other visual attributes of a real face, such as head orientation, even though this and other attributes are not labeled in either real or synthetic domain. Since our model learns to manipulate a specific property in isolation using only "synthetic demonstrations" of such manipulations without explicitly provided labels, it can be applied to shape, texture, lighting, and other properties that are difficult to measure or represent as real-valued vectors. We measure the degree to which our model preserves other attributes of a real image when a single specific attribute is manipulated. We use digit datasets to analyze how discrepancy in attribute distributions affects the performance of our model, and demonstrate results in a far more difficult setting: learning to manipulate real human faces using nonphotorealistic 3D renders. disentangled embeddings mouth head yaw, gaze, etc. discarded discarded E E E E G B G A mouth head yaw, gaze, etc. discarded discarded Model Modified Data Attribute J syn attr J syn rest J gen attr J gen rest
Distribution alignment has many applications in deep learning, including domain adaptation and unsupervised image-to-image translation. Most prior work on unsupervised distribution alignment relies either on minimizing simple non-parametric statistical distances such as maximum mean discrepancy, or on adversarial alignment. However, the former fails to capture the structure of complex real-world distributions, while the latter is difficult to train and does not provide any universal convergence guarantees or automatic quantitative validation procedures. In this paper we propose a new distribution alignment method based on a log-likelihood ratio statistic and normalizing flows. We show that, under certain assumptions, this combination yields a deep neural likelihoodbased minimization objective that attains a known lower bound upon convergence. We experimentally verify that minimizing the resulting objective results in domain alignment that preserves the local structure of input domains.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.