Mark Thompson

Simon Pochinda, Momen K. Tageldeen, Mark Thompson et al.

The increasing complexity of content rendering in modern games has led to a problematic growth in the workload of the GPU. In this paper, we propose an AI-based low-complexity scaler (LCS) inspired by state-of-the-art efficient super-resolution (ESR) models which could offload the workload on the GPU to a low-power device such as a neural processing unit (NPU). The LCS is trained on GameIR image pairs natively rendered at low and high resolution. We utilize adversarial training to encourage reconstruction of perceptually important details, and apply reparameterization and quantization techniques to reduce model complexity and size. In our comparative analysis we evaluate the LCS alongside the publicly available AMD hardware-based Edge Adaptive Scaling Function (EASF) and AMD FidelityFX Super Resolution 1 (FSR1) on five different metrics, and find that the LCS achieves better perceptual quality, demonstrating the potential of ESR models for upscaling on resource-constrained devices.

Srikanth Jonnada, Ram Dantu, Ishan Ranasinghe et al.

With the increasing ubiquity of technology in our daily lives, the complexity of our environment and the mechanisms required to function have also increased exponentially. Failure of any of the mechanical and digital devices that we rely on can be extremely disruptive. At times, the presence of an expert is needed to analyze, troubleshoot, and fix the problem. The increased demand and rapidly evolving mechanisms have led to an insufficient amount of skilled workers, thus resulting in long waiting times for consumers, and correspondingly high prices for expert services. We assert that performing a repair task with the guidance of experts from any geographical location provides an appropriate solution to the growing demand for handyman skills. This paper proposes an innovative mechanism for two geographically separated people to collaborate on a physical task. It also offers novel methods to analyze the efficiency of a collaboration system and a collaboration protocol through complexity indices. Using the innovative Collaborative Appliance for Remote-help (CARE) and with the support of a remote expert, fifty-nine subjects with minimal or no prior mechanical knowledge were able to elevate a car for replacing a tire; in a second experiment, thirty subjects with minimal or no prior plumbing knowledge were able to change the cartridge of a faucet. In both cases, average times were close to standard average repair times, and more importantly, both tasks were completed with total accuracy. Our experiments and results show that one can use the developed mechanism and methods for expanding the protocols for a variety of home, vehicle, and appliance repairs and installations.