How to use U^2-net in Tensorflow?

See original GitHub issue

I am trying to use the u2netp_480x640_float32.pb model in Tensorflow 2.3.1 (through the Java bindings but that should not make much of a difference). However I am not sure if my use of that model is incorrect or maybe something else is wrong.

What I do: load a 640x480 RGB image into a [1, 480, 640, 3] tensor and I scale the values between -1 and 1. I feed the tensor to the inputs layer and I fetch from the Identity layer, which is [1, 480, 640, 1]

What I get: U2Net01-2020-10-29-19 14 19

I also tried to scale values between 0-1, 0-255, -127-127. The latter gives me: U2Net01-2020-10-29-19 09 30

But that is not what I expect U^2-net to do either.

Am I missing something, or am I reading from the wrong layers?

Thanks

Issue Analytics

State:
Created 3 years ago
Comments:7 (5 by maintainers)

Top GitHub Comments

1reaction

PINTO0309commented, Oct 30, 2020

@edwinRNDR It depends on the properties of the model and the properties of the framework before and after the conversion. My goal in working on the model transformation is to do integer quantization in Tensorflow Lite, but TFLite does not currently allow for variable input resolution. Without that constraint, I would prefer to generate the model with variable size inputs. I have been forced to keep the input resolution fixed in order to centralize the model conversion workflow.

1reaction

PINTO0309commented, Oct 30, 2020

@edwinRNDR Thank you for the validation. There seems to be a mistake in the conversion of the model, so I will review it. Please give me some time.

Top Results From Across the Web

patrickisroman/u2net-tensorflow - GitHub

A tensorflow implementation of the U2-Net: Going Deeper with Nested U-Structure for Salient Object Detection using Keras & Functional API.

Usage of U2Net Model in android - pytorch - Stack Overflow

Hey , the current version of code at their github which is tensorflow examples runs smooth . They have optimised i guess ....

U2Net : A machine learning model that performs object ...

U2Net is a machine learning model that allows you to crop objects in a single shot. Taking an image of a person, cat,...

Image Background Removal with U^2-Net and OpenVINO

This notebook demonstrates background removal in images using U2-Net and ... display from model.u2net import U2NET, U2NETP from openvino.runtime import Core

U$^2$-Net: Going Deeper with Nested U-Structure for Salient ...

Task Dataset Model Metric Name Metric Value Glo... Dichotomous Image Segmentation DIS‑TE1 U2Net max F‑Measure 0.694 # 5 Dichotomous Image Segmentation DIS‑TE1 U2Net weighted F‑measure...