No output from UMT

[aweussom]

UMT appears to work fine, but no images are generated.
Ubuntu 20.04, x86_64

INITIALIZING UMT...
THRESHOLD: 0.5
CUSTOM LABEL MAP = FALSE
TPU = FALSE
CUSTOM DETECTOR = FALSE
INFO: Created TensorFlow Lite XNNPACK delegate for CPU.

TRACKING...
FRAME: 0
FRAME: 1
FRAME: 2
FRAME: 3
FRAME: 4
FRAME: 5
FRAME: 6
FRAME: 7
FRAME: 8
FRAME: 9

[aweussom]

Just testet from Docker on Raspberry Pi: Same problem. No images generated.

INITIALIZING UMT...
THRESHOLD: 0.5
CUSTOM LABEL MAP = FALSE
TPU = FALSE
CUSTOM DETECTOR = FALSE

TRACKING...
FRAME: 0
2022-05-24 14:50:46.707008: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)
2022-05-24 14:50:46.736980: W tensorflow/core/platform/profile_utils/cpu_utils.cc:116] Failed to find bogomips or clock in /proc/cpuinfo; cannot determine CPU frequency
FRAME: 1
FRAME: 2
FRAME: 3
FRAME: 4
FRAME: 5
FRAME: 6
FRAME: 7
FRAME: 8
FRAME: 9

[senker]

Just testet from Docker on Raspberry Pi: Same problem. No images generated.

INITIALIZING UMT...
THRESHOLD: 0.5
CUSTOM LABEL MAP = FALSE
TPU = FALSE
CUSTOM DETECTOR = FALSE

TRACKING...
FRAME: 0
2022-05-24 14:50:46.707008: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)
2022-05-24 14:50:46.736980: W tensorflow/core/platform/profile_utils/cpu_utils.cc:116] Failed to find bogomips or clock in /proc/cpuinfo; cannot determine CPU frequency
FRAME: 1
FRAME: 2
FRAME: 3
FRAME: 4
FRAME: 5
FRAME: 6
FRAME: 7
FRAME: 8
FRAME: 9

I guess I have the same problem now, but however it seems like I have some extra lines compared to what I can see from your side. Have you managed to find a solution for it?
My output:

WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation.
WARNING:root:Limited tf.summary API due to missing TensorBoard installation.

INITIALIZING UMT...
THRESHOLD: 0.5
CUSTOM LABEL MAP = FALSE
TPU = FALSE
CUSTOM DETECTOR = FALSE

TRACKING...
FRAME: 0
2022-05-31 09:08:23.012663: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)
2022-05-31 09:08:23.274399: W tensorflow/core/platform/profile_utils/cpu_utils.cc:116] Failed to find bogomips or clock in /proc/cpuinfo; cannot determine CPU frequency
2022-05-31 09:08:23.437208: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 8388608 exceeds 10% of free system memory.
2022-05-31 09:08:23.591266: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 8388608 exceeds 10% of free system memory.
2022-05-31 09:08:23.657290: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 8388608 exceeds 10% of free system memory.
2022-05-31 09:08:23.685661: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 8388608 exceeds 10% of free system memory.
2022-05-31 09:08:23.798316: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 8388608 exceeds 10% of free system memory.
FRAME: 1
FRAME: 2
FRAME: 3
FRAME: 4
FRAME: 5
FRAME: 6
FRAME: 7
FRAME: 8
FRAME: 9

[aweussom]

I guess I have the same problem now, but however it seems like I have some extra lines compared to what I can see from your side. Have you managed to find a solution for it? My output:

WARNING:root:Limited tf.compat.v2.summary API due to missing TensorBoard installation. WARNING:root:Limited
[...]

2022-05-31 09:08:23.798316: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 8388608 exceeds 10% of free system memory.
FRAME: 1
[...]
FRAME: 9

My output was the same as yours, I just trimmed it down a bit.

[rdpickard]

There's a bug in the documentation. The command needs the "-save" flag

umt -save -video highway_01.mp4

The detection didn't work, but the frames were put into an 'output' directory.