Ramana Sundararaman

I find an ambiguity in the way hstart, hend, wstart, wend are computed. When the pooling height and width are (3,3), the variable bin_size_h and bin_size_w are `2.667` (approximated to...

A way to apply SfM from frames belonging to Multi-camera with shared FoV?

1

Hi, I am trying to understand whether it is possible to run SfM for a scene involving moving targets, but static multi-view cameras. (Initial pose of the camera wouldn't change...

[snap] could not find "wayland-egl"

4

Hi, I am facing post-installation issues. 1. Installed without any errors using snap, `snap install cloudcompare`. 2. Tried to run cloudcompare, cloudcompare.CloudCompare Observation ``` This application failed to start because...

Implemented image coordinate metrics

1

Hello Chirstoph Heindl, As per our email correspondences, here is the pull request implementing the IDEucl metric proposed in our paper [Tracking Pedestrian Heads in Dense crowds](https://arxiv.org/abs/2103.13516) Thank you

Detector perf

13

Hello Christoph, I have added the MODP and the MODA metric in this PR, implemented from this paper : https://catalog.ldc.upenn.edu/docs/LDC2011V03/ClearEval_Protocol_v5.pdf. I however didn't understand the difference between MOC and MODA...

Modification of RNN doc

2

Addresses #812. Edit : Will push a commit for second part of the issue soon

Recently, I have been working with DAEs and found the problem of over-smoothness in denoised images(as I was using MSE as loss function). There have been recent works[1] on replacing...

Making custom input variables to InputLayer broadcastable

5

Continuation of https://github.com/Lasagne/Lasagne/pull/633

Raise error while trying to concat two Layers that are unequal

3

When I try to merge two Layers that have different dimension along the axis that is not in consideration for merging, I don't get an error. However the error is...

Porting some Layers from Gym

7

There are some layers in OpenAI [Gym](https://github.com/openai/improved-gan/blob/master/mnist_svhn_cifar10/nn.py) that would be helpful in training GANs. They're all Lasagne layers and it would be great to have them here directly. Namely, 1)...