4D reconstruction of human-object interaction is critical for immersive VR/AR experience and human activity understanding. Recent advances still fail to recover fine geometry and texture results from sparse RGB inputs, especially under challenging hu
man-object interactions scenarios. In this paper, we propose a neural human performance capture and rendering system to generate both high-quality geometry and photo-realistic texture of both human and objects under challenging interaction scenarios in arbitrary novel views, from only sparse RGB streams. To deal with complex occlusions raised by human-object interactions, we adopt a layer-wise scene decoupling strategy and perform volumetric reconstruction and neural rendering of the human and object. Specifically, for geometry reconstruction, we propose an interaction-aware human-object capture scheme that jointly considers the human reconstruction and object reconstruction with their correlations. Occlusion-aware human reconstruction and robust human-aware object tracking are proposed for consistent 4D human-object dynamic reconstruction. For neural texture rendering, we propose a layer-wise human-object rendering scheme, which combines direction-aware neural blending weight learning and spatial-temporal texture completion to provide high-resolution and photo-realistic texture results in the occluded scenarios. Extensive experiments demonstrate the effectiveness of our approach to achieve high-quality geometry and texture reconstruction in free viewpoints for challenging human-object interactions.
We present a study of the transport properties of thermally generated spin currents in an insulating ferrimagnetic-antiferromagnetic-ferrimagnetic trilayer over a wide range of temperature. Spin currents generated by the spin Seebeck effect (SSE) in
a yttrium iron garnet (YIG) YIG/NiO/YIG trilayer on a gadolinium gallium garnet (GGG) substrate were detected using the inverse spin Hall effect in Pt. By studying samples with different NiO thicknesses, the NiO spin diffusion length was determined to be 4.2 nm at room temperature. Interestingly, below 30 K, the inverse spin Hall signals are associated with the GGG substrate. The field dependence of the signal follows a Brillouin function for a S=7/2 spin ($mathrm{Gd^{3+}}$) at low temperature. Sharp changes in the SSE signal at low fields are due to switching of the YIG magnetization. A broad peak in the SSE response was observed around 100 K, which we associate with an increase in the spin-diffusion length in YIG. These observations are important in understanding the generation and transport properties of spin currents through magnetic insulators and the role of a paramagnetic substrate in spin current generation.
We present measurements of the spin Seebeck effect (SSE) by a technique that combines alternating currents (AC) and direct currents (DC). The method is applied to a ferrimagnetic insulator/heavy metal bilayer, Y$_3$Fe$_5$O$_{12}$(YIG)/Pt. Typically,
SSE measurements use an AC current to produce an alternating temperature gradient and measure the voltage generated by the inverse spin-Hall effect in the heavy metal at twice the AC frequency. Here we show that when Joule heating is associated with AC and DC bias currents, the SSE response occurs at the frequency of the AC current drive and can be larger than the second harmonic SSE response. We compare the first and second harmonic responses and show that they are consistent with the SSE. The field dependence of the voltage response is used to characterize the damping-like and field-like torques. This method can be used to explore nonlinear thermoelectric effects and spin dynamics induced by temperature gradients.
