Radiance Field methods have recently revolutionized novel-view synthesis of scenes captured with multiple photos or videos. However, achieving high visual quality still requires neural networks that are costly to train and render, while recent faster methods i...

Recent text-to-image generative models have demonstrated an unparalleled ability to generate diverse and creative imagery guided by a target text prompt. While revolutionary, current state-of-the-art diffusion models may still fail in generating images that fu...

The tremendous progress in neural image generation, coupled with the emergence of seemingly omnipotent vision-language models has finally enabled text-based interfaces for creating and editing images. Handling generic images requires a diverse underlying gener...

The construction industry is facing enormous pressure to adopt digital solutions to solve the industry's inherent problems. The digital twin has emerged as a solution that can update a BIM model with real-time data to achieve cyber-physical integration, enabli...

Diffusion models have achieved promising results in image restoration tasks, yet suffer from time-consuming, excessive computational resource consumption, and unstable restoration. To address these issues, we propose a robust and efficient Diffusion-based Low-...

Neural radiance fields enable state-of-the-art photorealistic view synthesis. However, existing radiance field representations are either too compute-intensive for real-time rendering or require too much memory to scale to large scenes. We present a Memory-Eff...

Many challenges to operate digital twins remain, hindering their design and implementation, and are rarely discussed. Furthermore, issues of social and legal nature are often overlooked. We identify the challenges of operating digital twins in the urban contex...

Diffusion models have experienced a surge of interest as highly expressive yet efficiently trainable probabilistic models. We show that these models are an excellent fit for synthesising human motion that co-occurs with audio, e.g., dancing and co-speech gesti...

Text-to-image personalization aims to teach a pre-trained diffusion model to reason about novel, user provided concepts, embedding them into new scenes guided by natural language prompts. However, current personalization approaches struggle with lengthy traini...

The automatic generation of stylized co-speech gestures has recently received increasing attention. Previous systems typically allow style control via predefined text labels or example motion clips, which are often not flexible enough to convey user intent acc...

The implementation of emerging technologies generates new possibility for the Architectural Engineering and Construction sector. Specifically, the study focuses on the applications of Digital Twins in the heritage construction sector, an area characterized by ...

Representing human performance at high-fidelity is an essential building block in diverse applications, such as film production, computer games or videoconferencing. To close the gap to production-level quality, we introduce HumanRF 1 , a 4D dynamic neural sce...