A survey on monocular 3D human pose estimation

Recovering human pose from RGB images and videos has drawn increasing attention in recent years owing to minimum sensor requirements and applicability in diverse fields such as human-computer interaction,robotics,video analytics,and augmented reality.Although a large amount of work has been devoted to this field,3D human pose estimation based on monocular images or videos remains a very challenging task due to a variety of difficulties such as depth ambiguities,occlusion,background clutters,and lack of training data.In this survey,we summarize recent advances in monocular 3D human pose estimation.We provide a general taxonomy to cover existing approaches and analyze their capabilities and limitations.We also present a summary of extensively used datasets and metrics,and provide a quantitative comparison of some representative methods.Finally,we conclude with a discussion on realistic challenges and open problems for future research directions.

A Domestication-Associated Gene GmPRR3b Regulates the Circadian Clock and Flowering Time in Soybean

Improved soybean cultivars have been adapted to grow at a wide range of latitudes,enabling expansion of cultivation worldwide.However,the genetic basis of this broad adaptation is still not clear.Here,we report the identification of GmPRR3b as a major flowering time regulatory gene that has been selected during domestication and genetic improvement for geographic expansion.Through a genome-wide association study of a diverse soybean landrace panel consisting of 279 accessions,we identified 16 candidate quantitative loci associated with flowering time and maturity time.The strongest signal resides in the known flowering gene E2,verifying the effectiveness of our approach.We detected strong signals associated with both flowering and maturity time in a genomic region containing GmPRR3b.Haplotype analysis revealed that GmPRR3bH6 is the major form of GmPRR3b that has been utilized during recent breeding of modern cultivars.mRNA profiling analysis showed that GmPRR3bH6 displays rhythmic and photoperiod-dependent expression and is preferentially induced under long-day conditions.Overexpression of GmPRR3bH6 increased main stem node number and yield,while knockout of GmPRR3bH6 using CRISPR/Cas9 technology delayed growth and the floral transition.GmPRR3bH6 appears to act as a transcriptional repressor of multiple predicted circadian clock genes,including GmCCAIa,which directly upregulates J/GmELF3a to modulate flowering time.The causal SNP(Chr12:5520945)likely endows GmPRR3bH6 a moderate but appropriate level of activity,leading to early flowering and vigorous growth traits preferentially selected during broad adaptation of landraces and improvement of cultivars.

Towards efficient and photorealistic 3D human reconstruction: A brief survey

Reconstructing 3D digital models of humans from sensory data is a long-standing problem in computer vision and graphics with a variety of applications in VR/AR,film production,and human–computer interaction,etc.While a huge amount of effort has been devoted to developing various capture hardware and reconstruction algorithms,traditional reconstruction pipelines may still suffer from high-cost capture systems and tedious capture processes,which prevent them from being easily accessible.Moreover,the dedicatedly hand-crafted pipelines are prone to reconstruction artifacts,resulting in limited visual quality.To solve these challenges,the recent trend in this area is to use deep neural networks to improve reconstruction efficiency and robustness by learning human priors from existing data.Neural network-based implicit functions have been also shown to be a favorable 3D representation compared to traditional forms like meshes and voxels.Furthermore,neural rendering has emerged as a powerful tool to achieve highly photorealistic modeling and re-rendering of humans by end-to-end optimizing the visual quality of output images.In this article,we will briefly review these advances in this fast-developing field,discuss the advantages and limitations of different approaches,and finally,share some thoughts on future research directions.

Optical manipulation of ratio-designable Janus microspheres

Angular optical trapping based on Janus microspheres has been proven to be a novel method to achieve controllable rotation.In contrast to natural birefringent crystals,Janus microspheres are chemically synthesized of two compositions with different refractive indices.Thus,their structures can be artificially regulated,which brings excellent potential for fine and multi-degree-of-freedom manipulation in the optical field.However,it is a considerable challenge to model the interaction of heterogeneous particles with the optical field,and there has also been no experimental study on the optical manipulation of microspheres with such designable refractive index distributions.How the specific structure affects the kinematic properties of Janus microspheres remains unknown.Here,we report systematic research on the optical trapping and rotating of various ratio-designable Janus microspheres.We employ an efficient T-matrix method to rapidly calculate the optical force and torque on Janus microspheres to obtain their trapped postures and rotational characteristics in the optical field.We have developed a robust microfluidic-based scheme to prepare Janus microspheres.Our experimental results demonstrate that within a specific ratio range,the rotation radii of microspheres vary linearly and the orientations of microsphere are always aligned with the light polarization direction.This is of great importance in guiding the design of Janus microspheres.And their orientations flip at a particular ratio,all consistent with the simulations.Our work provides a reliable theoretical analysis and experimental strategy for studying the interaction of heterogeneous particles with the optical field and further expands the diverse manipulation capabilities of optical tweezers.