rcnn-optimize

rfcn/faster rcnn optimize

rpn: C code
roi: multi-thread C code
Vector * Vector :avx test
level_thread: multi-thread matrix multiple sgemm demo form openblas
ristretto_rfcn: R-FCN 自动定点化工具实现，将ristretto移植进入faster rcnn框架下
A10_FPGA_Dma: Altera 10 FPGA DMA驱动
rfcn_net: rfcn非卷积部分实现(softmax rpn psroi ave_pooling)以及最后的box regress, nms处理
rpn_optimize: 对rpn层的优化,算法改进等处理，性能提高20~100倍
net_modle_optimize: 对基础ZF Net网络模型的优化,借鉴Inception、shuffle net、Crelu、dw conv等等结构，性能提升2~8倍

Name		Name	Last commit message	Last commit date
Latest commit History 101 Commits
A10_FPGA_Dma		A10_FPGA_Dma
ave_pool/single_thread		ave_pool/single_thread
ffmpeg_rtsp_inte_media_sdk		ffmpeg_rtsp_inte_media_sdk
intel_avx		intel_avx
level_thread		level_thread
matrix_transpose_see		matrix_transpose_see
multithreading_roi		multithreading_roi
net_modle_optimize/zf-net-rfcn		net_modle_optimize/zf-net-rfcn
nhwc_2_nchw		nhwc_2_nchw
numpy_shape_optimize/img_to_blob		numpy_shape_optimize/img_to_blob
optimize_gemm_step_by_step		optimize_gemm_step_by_step
psroi		psroi
pthread_fc_cache_and_sharejob		pthread_fc_cache_and_sharejob
rf_roi		rf_roi
rfcn_net		rfcn_net
ristretto_rfcn		ristretto_rfcn
roi_pooling_python		roi_pooling_python
rpclib		rpclib
rpn_optimize		rpn_optimize
rpn_proposal		rpn_proposal
softmax		softmax
sort		sort
README.md		README.md