After entering the command to text detection demo cpp with text-recognition-0016 model :
./text_detection_demo -i 1.png -m_td text_detection/text.xml
-m_tr text_recog_ori_32/text-recognition-0016-encoder.xml -tr_pt_first
-m_tr_ss "?0123456789abcdefghijklmnopqrstuvwxyz"
-tr_o_blb_nm "logits" -dt simple -lower
i got this error :
[ INFO ] OpenVINO
[ INFO ] version: 2022.1.0
[ INFO ] build: 2022.1.0-7019-cdb9bec7210-releases/2022/1
[ INFO ] Reading model: text_detection/text.xml
[ INFO ] Model name: saved_model
[ INFO ] Inputs:
[ INFO ] Cast_1:0, f32, {1,3,512,512}, [...]
[ INFO ] Outputs:
[ INFO ] detections, f32, {1,1,100,7}, [...]
[ INFO ] Preprocessor configuration:
[ INFO ] Input "Cast_1:0":
User's input tensor: {1,512,512,3}, [N,H,W,C], u8
Model's expected tensor: {1,3,512,512}, [N,C,H,W], f32
Implicit pre-processing steps (2):
convert type (f32): ({1,512,512,3}, [N,H,W,C], u8) -> ({1,512,512,3}, [N,H,W,C], f32)
convert layout [N,C,H,W]: ({1,512,512,3}, [N,H,W,C], f32) -> ({1,3,512,512}, [N,C,H,W], f32)
[ INFO ] The Text Detection model text_detection/text.xml is loaded to CPU
[ INFO ] Device: CPU
[ INFO ] Number of streams: 1
[ INFO ] Number of threads: AUTO
[ INFO ] Reading model: text_recog_ori_32/text-recognition-0016-encoder.xml
[ INFO ] Model name: torch-jit-export
[ INFO ] Inputs:
[ INFO ] imgs, f32, {1,1,64,256}, [N,C,H,W]
[ INFO ] Outputs:
[ INFO ] features, f32, {1,36,1024}, [...]
[ INFO ] decoder_hidden, f32, {1,1,1024}, [...]
[ INFO ] Preprocessor configuration:
[ INFO ] Input "imgs":
User's input tensor: {1,64,256,1}, [N,H,W,C], u8
Model's expected tensor: {1,1,64,256}, [N,C,H,W], f32
Implicit pre-processing steps (2):
convert type (f32): ({1,64,256,1}, [N,H,W,C], u8) -> ({1,64,256,1}, [N,H,W,C], f32)
convert layout [N,C,H,W]: ({1,64,256,1}, [N,H,W,C], f32) -> ({1,1,64,256}, [N,C,H,W], f32)
[ INFO ] The Composite Text Recognition model text_recog_ori_32/text-recognition-0016-encoder.xml is loaded to CPU
[ INFO ] Device: CPU
[ INFO ] Number of streams: 1
[ INFO ] Number of threads: AUTO
[ INFO ] Reading model: text_recog_ori_32/text-recognition-0016-decoder.xml
[ INFO ] Model name: torch-jit-export
[ INFO ] Inputs:
[ INFO ] hidden, f32, {1,1,1024}, [N,C,D]
[ INFO ] features, f32, {1,36,1024}, [N,C,D]
[ INFO ] decoder_input, f32, {1}, [N]
[ INFO ] Outputs:
[ INFO ] decoder_hidden, f32, {1,1,1024}, [...]
[ INFO ] decoder_output, f32, {1,40}, [...]
[ INFO ] The Composite Text Recognition model text_recog_ori_32/text-recognition-0016-decoder.xml is loaded to CPU
[ INFO ] Device: CPU
[ INFO ] Number of streams: 1
[ INFO ] Number of threads: AUTO
[ ERROR ] Failed to determine output blob names
Please clarify this error.
