realtime_segmenation icon indicating copy to clipboard operation
realtime_segmenation copied to clipboard

Realtime Semantic Segmentation for Mobile Platforms based on Tensorflow's Deeblap Model

Results 8 realtime_segmenation issues
Sort by recently updated
recently updated
newest added

Hi, Could you review how you implement bounding boxes with deeplab? I am using deeplab on coral and trying to do a similar thing, and I could not fully understand...

Hi @GustavZ Really cool work! Is there a way to just segment one class, i.e., just a person, instead of detecting various objects in the scene?

Can you share any information to train the model for road segmentation

I am trying to convert this model to coreML model using tfcoreml, however I'm getting errors. can someone guide me please?

Hi @GustavZ ! I tested run.py (changed video_input) using TF 1.3, and got this problem: InvalidArgumentError (see above for traceback): NodeDef mentions attr 'dilations' not in Op output:T; attr=T:type,allowed=[DT_HALF, DT_FLOAT];...

Hi, it's a impressive work, can I understand it as a derivative of semantic segmentation? After segment possible regions, you draw bounding boxes onto classes neither background nor small area,...

bazel-bin/tensorflow/contrib/lite/toco/toco --input_file=frozen_inference_graph.pb --input_format=TENSORFLOW_GRAPHDEF --output_format=TFLITE --output_file=real_segmenation.tflite --inference_type=FLOAT --inference_input_type=FLOAT --input_shape=1,368,432,3 --input_array=ImageTensor --output_array=SemanticPredictions result: Check failed: output_size_shape.dimensions_count() == 1 (2 vs. 1)