Inference error on Jetson: failed to load image

I run according to the object detection demo at:
https://github.com/dusty-nv/jetson-inference#locating-object-coordinates-using-detectnet

I’m running detectnet-console on a network I trained on my host pc, and run it on an image that exists.
I had one good run. When I try to run again, it keeps failing with the error:
failed to load image.

Any idea why?

Thanks!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
nvidia@tegra-ubuntu:~/DetectNet-COCO-Dog1-010818-1109-100epochs$ /home/nvidia/jetson-inference/build/aarch64/bin/detectnet-console 007414.jpg output007414.jpg --prototxt=$NET/deploy.prototxt --model=$NET/snapshot_iter_38600.caffemodel --input_blob=data --output_cvg=coverage --output_bbox=bboxes
detectnet-console
  args (8):  0 [/home/nvidia/jetson-inference/build/aarch64/bin/detectnet-console]  1 [007414.jpg]  2 [output007414.jpg]  3 [--prototxt=/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/deploy.prototxt]  4 [--model=/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel]  5 [--input_blob=data]  6 [--output_cvg=coverage]  7 [--output_bbox=bboxes]  


detectNet -- loading detection network model from:
          -- prototxt    /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/deploy.prototxt
          -- model       /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel
          -- input_blob  'data'
          -- output_cvg  'coverage'
          -- output_bbox 'bboxes'
          -- mean_pixel  0.000000
          -- threshold   0.500000
          -- batch_size  2

[GIE]  TensorRT version 3.0.4, build 3004
[GIE]  attempting to open cache file /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel.2.tensorcache
[GIE]  cache file not found, profiling network model
[GIE]  platform has FP16 support.
[GIE]  loading /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/deploy.prototxt /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel
[GIE]  retrieved output tensor 'coverage'
[GIE]  retrieved output tensor 'bboxes'
[GIE]  configuring CUDA engine
[GIE]  building CUDA engine
[GIE]  completed building CUDA engine
[GIE]  network profiling complete, writing cache to /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel.2.tensorcache
[GIE]  completed writing cache to /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel.2.tensorcache
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel loaded
[GIE]  CUDA engine context initialized with 3 bindings
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel input  binding index:  0
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel input  dims (b=2 c=3 h=640 w=640) size=9830400
[cuda]  cudaAllocMapped 9830400 bytes, CPU 0x101540000 GPU 0x101540000
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel output 0 coverage  binding index:  1
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel output 0 coverage  dims (b=2 c=1 h=40 w=40) size=12800
[cuda]  cudaAllocMapped 12800 bytes, CPU 0x101ea0000 GPU 0x101ea0000
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel output 1 bboxes  binding index:  2
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel output 1 bboxes  dims (b=2 c=4 h=40 w=40) size=51200
[cuda]  cudaAllocMapped 51200 bytes, CPU 0x1020a0000 GPU 0x1020a0000
/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel initialized.
[cuda]  cudaAllocMapped 16 bytes, CPU 0x101340200 GPU 0x101340200
maximum bounding boxes:  6400
[cuda]  cudaAllocMapped 102400 bytes, CPU 0x1022a0000 GPU 0x1022a0000
[cuda]  cudaAllocMapped 25600 bytes, CPU 0x1020ac800 GPU 0x1020ac800
failed to load image 007414.jpg
failed to load image '007414.jpg'
~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Hi zvika.peter, can you try changing your terminal’s working directory before launching the detectnet-console? Try this instead:

# after defining $NET environment variable to `~/DetectNet-COCO-Dog1-010818-1109-100epochs`
$ cd /home/nvidia/jetson-inference/build/aarch64/bin/
$ ./detectnet-console $NET/007414.jpg $NET/output007414.jpg --prototxt=$NET/deploy.prototxt --model=$NET/snapshot_iter_38600.caffemodel --input_blob=data --output_cvg=coverage --output_bbox=bboxes

Does that help?

Thanks duty_nv, but still not working…

Any more ideas why it cant find a file that exists?

nvidia@tegra-ubuntu:~/jetson-inference/build/aarch64/bin$ ./detectnet-console $NET/007414.jpg $NET/output007414.jpg --prototxt=$NET/deploy.prototxt --model=$NET/snapshot_iter_38600.caffemodel --input_blob=data --output_cvg=coverage --output_bbox=bboxes
detectnet-console
  args (8):  0 [./detectnet-console]  1 [/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/007414.jpg]  2 [/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/output007414.jpg]  3 [--prototxt=/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/deploy.prototxt]  4 [--model=/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel]  5 [--input_blob=data]  6 [--output_cvg=coverage]  7 [--output_bbox=bboxes]  


detectNet -- loading detection network model from:
          -- prototxt    /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/deploy.prototxt
          -- model       /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel
          -- input_blob  'data'
          -- output_cvg  'coverage'
          -- output_bbox 'bboxes'
          -- mean_pixel  0.000000
          -- threshold   0.500000
          -- batch_size  2

[GIE]  TensorRT version 3.0.4, build 3004
[GIE]  attempting to open cache file /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel.2.tensorcache
[GIE]  loading network profile from cache... /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel.2.tensorcache
[GIE]  platform has FP16 support.
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel loaded
[GIE]  CUDA engine context initialized with 3 bindings
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel input  binding index:  0
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel input  dims (b=2 c=3 h=640 w=640) size=9830400
[cuda]  cudaAllocMapped 9830400 bytes, CPU 0x101540000 GPU 0x101540000
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel output 0 coverage  binding index:  1
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel output 0 coverage  dims (b=2 c=1 h=40 w=40) size=12800
[cuda]  cudaAllocMapped 12800 bytes, CPU 0x101ea0000 GPU 0x101ea0000
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel output 1 bboxes  binding index:  2
[GIE]  /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel output 1 bboxes  dims (b=2 c=4 h=40 w=40) size=51200
[cuda]  cudaAllocMapped 51200 bytes, CPU 0x1020a0000 GPU 0x1020a0000
/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/snapshot_iter_38600.caffemodel initialized.
[cuda]  cudaAllocMapped 16 bytes, CPU 0x101340200 GPU 0x101340200
maximum bounding boxes:  6400
[cuda]  cudaAllocMapped 102400 bytes, CPU 0x1022a0000 GPU 0x1022a0000
[cuda]  cudaAllocMapped 25600 bytes, CPU 0x1020ac800 GPU 0x1020ac800
failed to load image /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/007414.jpg
failed to load image '/home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/007414.jpg'
nvidia@tegra-ubuntu:~/jetson-inference/build/aarch64/bin$ ls -l /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/007414.jpg
-rw-rw-r-- 1 nvidia nvidia 63610880 Aug  1 14:08 /home/nvidia/DetectNet-COCO-Dog1-010818-1109-100epochs/007414.jpg
nvidia@tegra-ubuntu:~/jetson-inference/build/aarch64/bin$

I just tried with another file and it worked.
Have no idea why (the file is not corrupted and not open by the os)
Closing issue.
Thanks for the support!