Key Pose Frame Extraction Method of Human Motion Based on 3D Framework and X-Means

The key pose frames of a human motion pose sequence,play an important role in the compression,retrieval and semantic analysis of continuous human motion.The current available clustering methods in literatures are difficult to determine the number of key pose frames automatically,and may destroy the postures’ temporal relationships while extracting key frames.To deal with this problem,this paper proposes a new key pose frames extraction method on the basis of 3D space distances of joint points and the improved X-means clustering algorithm.According to the proposed extraction method,the final key pose frame sequence could be obtained by describing the posture of human body with space distance of particular joint points and then the time-constraint X-mean algorithm is applied to cluster and filtrate the posture sequence.The experimental results show that the proposed method can automatically determine the number of key frames and save the temporal characteristics of motion frames according to the motion pose sequence.

Vehicle color recognition based on smooth modulation neural network with multi-scale feature fusion

Vehicle Color Recognition(VCR)plays a vital role in intelligent traffic management and criminal investigation assistance.However,the existing vehicle color datasets only cover 13 classes,which can not meet the current actual demand.Besides,although lots of efforts are devoted to VCR,they suffer from the problem of class imbalance in datasets.To address these challenges,in this paper,we propose a novel VCR method based on Smooth Modulation Neural Network with Multi-Scale Feature Fusion(SMNN-MSFF).Specifically,to construct the benchmark of model training and evaluation,we first present a new VCR dataset with 24 vehicle classes,Vehicle Color-24,consisting of 10091 vehicle images from a 100-hour urban road surveillance video.Then,to tackle the problem of long-tail distribution and improve the recognition performance,we propose the SMNN-MSFF model with multiscale feature fusion and smooth modulation.The former aims to extract feature information from local to global,and the latter could increase the loss of the images of tail class instances for training with class-imbalance.Finally,comprehensive experimental evaluation on Vehicle Color-24 and previously three representative datasets demonstrate that our proposed SMNN-MSFF outperformed state-of-the-art VCR methods.And extensive ablation studies also demonstrate that each module of our method is effective,especially,the smooth modulation efficiently help feature learning of the minority or tail classes.Vehicle Color-24 and the code of SMNN-MSFF are publicly available and can contact the author to obtain.