Conference paper Open Access

Attention-enhanced Sensorimotor Object Recognition

Thermos, S; Papadopoulos, GT; Daras, P; Potamianos, G


Citation Style Language JSON Export

{
  "DOI": "10.1109/ICIP.2018.8451158", 
  "language": "eng", 
  "title": "Attention-enhanced Sensorimotor Object Recognition", 
  "issued": {
    "date-parts": [
      [
        2018, 
        10, 
        10
      ]
    ]
  }, 
  "abstract": "<p>Sensorimotor learning, namely the process of understanding the physical world by combining visual and motor information, has been recently investigated, achieving promising results for the task of 2D/3D object recognition. Following the recent trend in computer vision, powerful deep neural networks (NNs) have been used to model the &ldquo;sensory&rdquo; and &ldquo;motor&rdquo; information, namely the object appearance and affordance. However, the existing implementations cannot efficiently address the spatio-temporal nature of the humanobject interaction. Inspired by recent work on attention-based learning, this paper introduces an attention-enhanced NN-based model that learns to selectively focus on parts of the physical interaction where the object appearance is corrupted by occlusions and deformations. The model&rsquo;s attention mechanism relies on the confidence of classifying an object based solely on its appearance. Three metrics are used to measure the latter, namely the prediction entropy, the average N-best likelihood difference, and the N-best likelihood dispersion. Evaluation of the attention-enhanced model on the SOR3D dataset reports 33% and 26% relative improvement over the appearance-only and the spatio-temporal fusion baseline models, respectively.</p>", 
  "author": [
    {
      "family": "Thermos, S"
    }, 
    {
      "family": "Papadopoulos, GT"
    }, 
    {
      "family": "Daras, P"
    }, 
    {
      "family": "Potamianos, G"
    }
  ], 
  "id": "3727849", 
  "version": "pre-print", 
  "type": "paper-conference", 
  "event": "IEEE ICIP 2018"
}
24
96
views
downloads
Views 24
Downloads 96
Data volume 74.7 MB
Unique views 21
Unique downloads 95

Share

Cite as