Does the relization of the Segnet in network.py match the paper of segnet?

[mlx123]

Hello I read the the relization of the Segnet in network.py and I have a question. In the paper of Segnet,the last layer boxed in the picture below is (one Upsampling + one Conv +one Conv)

but in the relization of the Segnet in network.py (https://github.com/bodokaiser/piwise/blob/50f8be658693dd27bfa463cb00d1acec7e869fc4/piwise/network.py#L275),the layer metioned above is (one Upsampling + one Conv) ,it lose one Conv layer.

Thank you

[bodokaiser]

So what is your question?

On 4. Jan 2019, at 17:03, LxM notifications@github.com wrote:

Hello I read the the relization of the Segnet in network.py and I have a question. In the paper of Segnet,the last layer boxed in the picture below is (one Upsampling + one Conv +one Conv) https://user-images.githubusercontent.com/28173246/50697383-06a5db00-107d-11e9-92ca-5bba69f8a965.png but in the relization of the Segnet in network.py (https://github.com/bodokaiser/piwise/blob/50f8be658693dd27bfa463cb00d1acec7e869fc4/piwise/network.py#L275),the https://github.com/bodokaiser/piwise/blob/50f8be658693dd27bfa463cb00d1acec7e869fc4/piwise/network.py#L275),the layer metioned above is (one Upsampling + one Conv) ,it lose one Conv layer. https://user-images.githubusercontent.com/28173246/50697496-58e6fc00-107d-11e9-98bb-36c731734303.png Thank you

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/bodokaiser/piwise/issues/27, or mute the thread https://github.com/notifications/unsubscribe-auth/ABsq8oRCX_tMgp2WEv5WUxqZccisE2ZMks5u_3thgaJpZM4Zp5RY.

[bodokaiser]

Ah sry, did not see the email subject.

Q: Does the relization of the Segnet in network.py match the paper of segnet? A: As you have noticed this seems not the case. I am not sure what impact this has on performance to be honest.

On 4. Jan 2019, at 17:30, Bodo Kaiser bodo.rene.kaiser@gmail.com wrote:

So what is your question?

On 4. Jan 2019, at 17:03, LxM <notifications@github.com mailto:notifications@github.com> wrote:

Hello I read the the relization of the Segnet in network.py and I have a question. In the paper of Segnet,the last layer boxed in the picture below is (one Upsampling + one Conv +one Conv) https://user-images.githubusercontent.com/28173246/50697383-06a5db00-107d-11e9-92ca-5bba69f8a965.png but in the relization of the Segnet in network.py (https://github.com/bodokaiser/piwise/blob/50f8be658693dd27bfa463cb00d1acec7e869fc4/piwise/network.py#L275),the https://github.com/bodokaiser/piwise/blob/50f8be658693dd27bfa463cb00d1acec7e869fc4/piwise/network.py#L275),the layer metioned above is (one Upsampling + one Conv) ,it lose one Conv layer. https://user-images.githubusercontent.com/28173246/50697496-58e6fc00-107d-11e9-98bb-36c731734303.png Thank you

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/bodokaiser/piwise/issues/27, or mute the thread https://github.com/notifications/unsubscribe-auth/ABsq8oRCX_tMgp2WEv5WUxqZccisE2ZMks5u_3thgaJpZM4Zp5RY.