文件名称：matlab代码30行-VCAR-CNN:VCAR神经网络
文件大小：14.19MB
文件格式：ZIP
更新时间：2024-06-15 09:36:08
系统开源
matlab代码 30行 Neural Network-Based Video Compression Artifact Reduction using Temporal Correlation and Sparsity Prior Predictions(IEEE Access, No.8, 2020). Neural Network-Based Video Compression Artifact Reduction using Temporal Correlation and Sparsity Prior Predictions, IEEE Access, DOI: 10.1109/ACCESS.2018.2876864. Dependencies Keras, TensorFlow, NumPy, Matlab, OpenCV, ... Get Started! 1. Preparation (1)使用编码工具(如HM)对视频进行编码, 编码参数可参见sample_parameters.cfg. 目前的设置: I帧的间隔为32, 即第0, 32, 64帧为I帧, 其余为帧; P
【文件预览】：
VCAR-CNN-main
----allmodel(345KB)
----models()
--------QP42-weights-30.40967.hdf5(2.62MB)
----output(9B)
----model_car.py(3KB)
----BRISQUE()
--------brisquescore.m(835B)
--------allmodel(345KB)
--------estimateggdparam.m(535B)
--------output(9B)
--------test_ind(428B)
--------svm-scale.exe(79KB)
--------dump(226B)
--------allrange(741B)
--------estimateaggdparam.m(543B)
--------brisque_feature.m(1KB)
--------testimage2.bmp(1.13MB)
--------readme.txt(6KB)
--------svm-predict.exe(105KB)
--------test_ind_scaled(456B)
--------testimage1.bmp(1.13MB)
----read_y.m(585B)
----test_ind(430B)
----svm-scale.exe(79KB)
----dump(226B)
----demo.py(3KB)
----README.md(1KB)
----coded_str()
--------Ballpass_D_qp42_dec_65F.yuv(9.28MB)
--------Ballpass_D_ori_65F.yuv(9.28MB)
--------Fourpeople_E_qp42_65F.bin(81KB)
--------Racehorses_D_qp42_65F.bin(54KB)
--------BallDrill_C_qp42_65F.bin(98KB)
--------BQMall_C_qp42_65F.bin(118KB)
--------BlowingB_D_qp42_69F.bin(43KB)
--------Ballpass_D_qp42_65F.bin(23KB)
----allrange(741B)
----__pycache__()
--------common.cpython-36.pyc(2KB)
--------model_car.cpython-36.pyc(4KB)
----frame_compensation.p(2KB)
----comp_psnr_3im.m(872B)
----svm-predict.exe(105KB)
----test_ind_scaled(458B)
----sample_parameters.cfg(10KB)