An error occurred while processing the template.
The following has evaluated to null or missing:
==> RimuoviOmbra  [in template "20101#20127#625815" at line 6, column 56]

Tip: If the failing expression is known to be legally refer to something that's sometimes null or missing, either specify a default value like myOptionalVar!myDefault, or use <#if myOptionalVar??>when-present<#else>when-missing</#if>. (These only cover the last step of the expression; to cover the whole expression, use parenthesis: (myOptionalVar.foo)!myDefault, (myOptionalVar.foo)??

FTL stack trace ("~" means nesting-related):
	- Failed at: #if getterUtil.getBoolean(RimuoviOmbr...  [in template "20101#20127#625815" at line 6, column 29]
1<#assign bgcolor = "#ffffff"> 
2<#if BackgroundColor.getData()?has_content> 
3    <#assign bgcolor = BackgroundColor.getData()> 
6<div class="header header-1 <#if getterUtil.getBoolean(RimuoviOmbra.getData())>no-shadow</#if>"> 
7    <div class="image" style="background-color:${bgcolor}"> 
8        <#if Image.getData()?? && Image.getData() != ""> 
9            <img src="${Image.getData()}" alt=""> 
10        </#if> 
11    </div> 
12      <div class="header-body"> 
13        <div class="container"> 
14          <div class="header-title">${Title.getData()}</div> 
15          <div class="row"> 
16            <div class="col-8"> 
17              <div class="header-text">${Abstract.getData()}</div> 
18            </div> 
19          </div> 
20           <#if Link.getData()?? && LinkLabel.getData()?? && LinkLabel.getData() != ""> 
21            <div class="header-action"><a class="btn btn-primary" href="${Link.getData()}">${LinkLabel.getData()}</a></div> 
22           </#if> 
23        </div> 
24    </div> 


To better interact with its environment, an intelligent system must be capable of identifying the items around it. In the context of a humanoid platform, the capability of detecting and recognizing objects is paramount for a smooth integration to real life scenarios. The goal of this project is therefore providing iCub with this capability only relying on the information coming from the event driven cameras mounted in its eyes. To do so, we use Spiking Neural Networks (SNN), in the attempt of building a purely event driven framework.

The results of our experiments show promising recognition capabilities on a dataset collected showing a bunch of objects to the robot cameras. The training of the network is carried out using a back-propagation technique which relies on surrogate gradients to approximate the spike differentiation. Using such technique we can count on GPU-based training library such as PyTorch. In the figures we can see a data sample, as well as the test accuracy during training. At the same time, the learning rules used in this project, map on neuromorphic hardware (such as Loihi), going towards a fully spiking implementation.

Spiking Neural Networks, PyTorch, Deep Learning, Surrogate gradient, GPU



[Research Project Page]-Gallery_Title