Neural hand reconstruction using an RGB image

Background This study presents a neural hand reconstruction method for monocular 3D hand pose and shape estimation.Methods Alternate to directly representing hand with 3D data,a novel UV position map is used to represent a hand pose and shape with 2D data that maps 3D hand surface points to 2D image space.Furthermore,an encoder-decoder neural network is proposed to infer such UV position map from a single image.To train this network with inadequate ground truth training pairs,we propose a novel MANOReg module that employs MANO model as a prior shape to constrain high dimensional space of the UV position map.Results The quantitative and qualitative experiments demonstrate the effectiveness of our UV position map representation and MANOReg module.

Effect of soil surface roughness on emergence rate and yield of mechanized direct-seeded rapeseed based on 3D laser scanning

The quality of seedbed after sowing such as soil surface roughness is one of the key factors affecting the seedling emergence of rapeseed,which ultimately affected crop yield.However,the effect of soil surface roughness on seedling emergence and yield of rapeseed is still unclear.In this study,field experiments at the experimental site of Jianli and Shayang were carried out.Three treatments were designed:relative slow(M1),medium(M2),and fast(M3)forward speed of the unit.Soil surface roughness measured by a 3D laser scanner,seed quantity of the seeder,emergence rate and yield of rapeseed were determined to investigate the soil surface roughness effect on emergence rate and yield of rapeseed.The results showed that as the forward speed of the unit increased,the compartment surface became rougher.Compared with the M1 and M2 treatments,soil surface roughness under the M3 treatment increased by 36.5%and 9.8%,respectively.The actual seed quantity of the seeder under different treatments ranged from 3806.56 to 4158.18 g/hm2.The average error rate of the actual and theoretical seed quantity was less than 5%,which met the operational quality requirements for seeding rapeseed crops.As the forward speed of the unit increased,the actual seed quantity of the seeder gradually increased while the emergence rate and yield of rapeseed decreased.The seed quantity under the M3 treatment increased by 6.9%and 4.7%,while the emergence rate of rapeseed decreased by 3.3%and 2.0%,and the yield decreased by 23.2%and 13.1%,compared with the M1 and M2 treatments,respectively.Correlation analysis indicated that emergence rate and yield of rapeseed were negatively influenced by soil surface roughness.Considering rapeseed emergence rate,seed yield,and economic benefits,the M1 treatment was recommended.But considering the factor that the M1 treatment may reduce the unit operation efficiency,and thus resulting in lower cost of production,M2 could be recommended in actual farming.The results of this study laid a theoretical foundation for analyzing the relationship between the seedbed surface quality and seedling emergence and yield.