LMR2018

Why is the GPU memory footprint doubled with Tensorrt acceleration？

1

Why is the GPU memory footprint doubled with Tensorrt acceleration

When I use this command, the FPS is only 17, much less than the 100 + you gave me, any idea why?

1

When I use this command, the FPS is only 17, much less than the 100 + you gave me, any idea why? 

I want to convert the video like this and save it as Npy:

1

I want to convert the video like this and save it as Npy: The file should contain an array with shape 15 x 2 x length. The first dimension (15)...

how to generate a numpy .npy file from DensePose?

Hello, I want to know the training data set of Hand model, can you provide it to me?

摄像头启动后不久就黑屏停止不动了，程序没有报错，我用官方的来跑是正常的，知道这是什么问题吗？

3

D/UseCaseAttachState: All use case: [androidx.camera.core.Preview-27cce9c0-8fa0-445c-a7e1-b01fa4accbdb265504916] for camera: 0 D/DeferrableSurface: New surface in use[total_surfaces=1, used_surfaces=1](androidx.camera.core.SurfaceRequest$2@f72883} use count+1, useCount=1 androidx.camera.core.SurfaceRequest$2@f72883 D/CaptureSession: Opening capture session. D/CaptureSession: Attempting to send capture request onConfigured Issuing...

After loading hand_tracking_mobile_gpu.binarypb model, the code will not move after this step, delete this model will report an error, how should I solve this problem? 

想知道换了自己训练模型后，如何修改其相应的class和输入输出张量配置？

3

想知道换了自己训练模型后，如何修改其相应的class和输入输出张量配置？ W/ncnn: overwrite existing custom layer type YoloV5Focus D/current use model: picodet-s-416.param W/ncnn: AAssetManager_open picodet-s-416.param failed AAssetManager_open picodet-s-416.bin failed D/ncnn: current use model: picodet-s-416 E/model: current_model I/System.out: 0 1 D/mali_winsys:...

tiny-pose安卓模型代码运行：Wrong broadcast type是什么错误？怎么解决？

1

A/Paddle-Lite: [F 11/ 1 18: 0:58.555 .../lite/kernels/host/elementwise_op_func.h:339 Update] Wrong broadcast type E/libc++abi: terminating with uncaught exception of type paddle::lite::PaddleLiteException: Paddle-Lite C++ Exception: [F 11/ 1 18: 0:58.555 .../lite/kernels/host/elementwise_op_func.h:339 Update] Wrong...

ch_PP-OCRv4_rec_svtr_large.yml训练导出的模型，使用predict_rec.py预测宽度比较大的图片时出现(InvalidArgument) Broadcast dimension mismatch

9

使用ch_PP-OCRv4_rec_svtr_large.yml训练的OCR识别模型，训练正常， 使用python tools/eval.py -c configs/rec/PP-OCRv4/ch_PP-OCRv4_rec_svtr_large.yml也是正常的， 用python tools/export_model.py -c configs/rec/PP-OCRv4/ch_PP-OCRv4_rec_svtr_large.yml 也是能成功导出模型的 用export_model.py导出的模型，使用python tools/infer/predict_rec.py预测宽度不太长的单行图片也是能正常预测的， 但是predict_rec.py预测宽度比较大的图片时出现：(InvalidArgument) Broadcast dimension mismatch错误 即使是训练、验证用的宽度比较大的图片也是出现这个错误，这个怎么解决？ 训练配置：image_shape: [3, 48, 320]， max_text_length: &max_text_length 50 `[2024/05/25 14:48:36] ppocr INFO: Traceback (most...