Closed shiner-chen closed 7 months ago
我从你的github上拿下来的代码,跑下来怎么只有40多帧呢?请问哪里设置不对? Loading model... sdk version: 1.5.2 (c6b7b351a@2023-08-23T15:28:22) driver version: 0.8.8 model input num: 1, output num: 3 model is NHWC input fmt model input height=640, width=640, channel=3 Loading model... sdk version: 1.5.2 (c6b7b351a@2023-08-23T15:28:22) driver version: 0.8.8 model input num: 1, output num: 3 model is NHWC input fmt model input height=640, width=640, channel=3 Loading model... sdk version: 1.5.2 (c6b7b351a@2023-08-23T15:28:22) driver version: 0.8.8 model input num: 1, output num: 3 model is NHWC input fmt model input height=640, width=640, channel=3 rgaapi version 1.9.1【4】 loadLabelName ./model/coco_80_labels_list.txt 120帧内平均帧率: 45.924225 fps/s 120帧内平均帧率: 46.189376 fps/s 120帧内平均帧率: 46.296296 fps/s
我已经调用performance.sh脚本设置了cpu和npu最高性能模式
我发现是opencv解码上不去,不推理的情况下解码也只有50多帧,请问你的opencv带了硬件解码加速吗?
重新编译opencv+支持硬解码的ffmpeg,推理速度上去了,能达到100帧了多了
我从你的github上拿下来的代码,跑下来怎么只有40多帧呢?请问哪里设置不对? Loading model... sdk version: 1.5.2 (c6b7b351a@2023-08-23T15:28:22) driver version: 0.8.8 model input num: 1, output num: 3 model is NHWC input fmt model input height=640, width=640, channel=3 Loading model... sdk version: 1.5.2 (c6b7b351a@2023-08-23T15:28:22) driver version: 0.8.8 model input num: 1, output num: 3 model is NHWC input fmt model input height=640, width=640, channel=3 Loading model... sdk version: 1.5.2 (c6b7b351a@2023-08-23T15:28:22) driver version: 0.8.8 model input num: 1, output num: 3 model is NHWC input fmt model input height=640, width=640, channel=3 rgaapi version 1.9.1【4】 loadLabelName ./model/coco_80_labels_list.txt 120帧内平均帧率: 45.924225 fps/s 120帧内平均帧率: 46.189376 fps/s 120帧内平均帧率: 46.296296 fps/s
我已经调用performance.sh脚本设置了cpu和npu最高性能模式