With the rise of virtual avatars in professional, educational, and recreational settings, this study investigates how different avatar types—varying in realism, gender, and identity—affect user perceptions of embodiment, acceptability, technostress, privacy, and preferences. Two studies were conducted with 42 participants in Study 1 and 40 in Study 2, including professionals and students with varying VR experiences. In Study 1, participants used pre-assigned avatars they could control during interactions. In Study 2, an interviewer used different avatars to interact with participants and assess their impact. Questionnaires and correlation analyses measured embodiment, technostress, privacy, and preference variations across contexts. Results showed that hyper-realistic avatars resembling the user enhanced perceived embodiment and credibility in professional and educational settings, while non-realistic avatars were preferred in recreational contexts, particularly when interacting with strangers. Technostress was generally low, though younger users were more sensitive to avatar appearance, and privacy concerns increased when avatars were controlled by others. Gender differences emerged, with women expressing more concern about appearance and men preferring same-gender avatars in professional environments. These findings highlight the need for VR platform designers to balance realism with user comfort and address privacy concerns to encourage broader adoption in professional and educational applications.

The real world exhibits rich structure and detail across many scales of observation. It is difficult, however, to capture and represent a broad spectrum of scales using ordinary images. We devise a novel paradigm for learning a representation that captures an orders-of-magnitude variety of scales from an unstructured collection of ordinary images. We treat this collection as a distribution of scale-space slices to be learned using adversarial training, and additionally enforce coherency across slices. Our approach relies on a multiscale generator with carefully injected procedural frequency content, which allows to interactively explore the emerging continuous scale space. Training across vastly different scales poses challenges regarding stability, which we tackle using a supervision scheme that involves careful sampling of scales. We show that our generator can be used as a multiscale generative model, and for reconstructions of scale spaces from unstructured patches. Significantly outperforming the state of the art, we demonstrate zoom-in factors of up to 256x at high quality and scale consistency.

With the rise of virtual avatars in professional, educational, and recreational settings, this study investigates how different avatar types, with variations in gender and identity affect user perceptions of embodiment, acceptability, technostress, privacy, and preferences. To examine these factors, two studies were conducted in controlled virtual meeting environments. In Study 1, participants controlled their own avatars, while in Study 2, an interviewer used various avatars. Questionnaires and correlation analyses were employed to assess the impact of avatar characteristics on perceived embodiment, technostress, privacy, and preferences across different contexts. Results showed that hyper-realistic avatars, especially those resembling the user, enhanced perceived embodiment and were rated higher for credibility in professional and educational contexts. In contrast, non-realistic avatars were favored in recreational settings, particularly when interacting with strangers. Technostress was generally low, though younger users exhibited more sensitivity to avatar appearance, and privacy concerns increased when avatars were used by others. Gender differences also emerged, with women expressing more concern about appearance and men preferring same-gender avatars in professional and educational settings. Balancing realism with user comfort and addressing privacy concerns are essential for broader adoption.

Animating human face images aims to synthesize a desired source identity in a natural-looking way mimicking a driving video's facial movements. In this context, Generative Adversarial Networks have demonstrated remarkable potential in real-time face reenactment using a single source image, yet are constrained by limited geometry consistency compared to graphic-based approaches. In this paper, we introduce Geometry-guided GAN for Face Animation (G3FA) to tackle this limitation. Our novel approach empowers the face animation model to incorporate 3D information using only 2D images, improving the image generation capabilities of the talking head synthesis model. We integrate inverse rendering techniques to extract 3D facial geometry properties, improving the feedback loop to the generator through a weighted average ensemble of discriminators. In our face reenactment model, we leverage 2D motion warping to capture motion dynamics along with orthogonal ray sampling and volume rendering techniques to produce the ultimate visual output. To evaluate the performance of our G3FA, we conducted comprehensive experiments using various evaluation protocols on VoxCeleb2 and TalkingHead benchmarks to demonstrate the effectiveness of our proposed framework compared to the state-of-the-art real-time face animation methods.