Early Cognitive Vision: Using Gestalt-Laws for Task-Dependent, Active Image-Processing

Wörgötter, Florentin; Krüger, Norbert; Pugeault, Nicolas; Calow, Dirk; Lappe, Markus; Pauwels, Karl; Hulle, Marc M. Van; Tan, Sovira; Johnston, Alan

doi:10.1023/b:naco.0000036817.38320.fe

Cited by 18 publications

(12 citation statements)

References 86 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gestalt laws ( Figure 1.2) describe which local visual features support perceptual grouping. We (as others, e.g., Wörgötter et al, 2004) assume that Gestalt properties are manifest in the wiring and therefore we included three Gestalt laws on the network level in order to aid object coding.…”

Section: Gestalt Rulesmentioning

confidence: 99%

Coding the presence of visual objects in a recurrent neural network of visual cortex

Zwickel¹,

Wachtler²,

Eckhorn³

2007

Biosystems

View full text Add to dashboard Cite

Section: Gestalt Rulesmentioning

confidence: 99%

Coding the presence of visual objects in a recurrent neural network of visual cortex

Zwickel¹,

Wachtler²,

Eckhorn³

2007

Biosystems

View full text Add to dashboard Cite

“…The term 'early cognitive vision' in contrast to 'early vision' was first introduced in [48] in the context of the ECOVISION project [12]. At the stage of Early Vision, computer vision has dealt to a large extent with different visual modalities separately and in many computer vision systems, one or more of the aforementioned aspects are processed (see, e.g., [29,32,42]).…”

Section: Early Vision and Early Cognitive Visionmentioning

confidence: 99%

“…In the past decade, also mid-level processes (which is also called 'early cognitive vision'-see, e.g., [48]) for complex feature extraction and contextual disambiguation have been a focus of interest realizing intermediate levels of human visual processing. It is important to note that these processes can have rather different implementational structures that in some cases make it less feasible to realize them on GPUs.…”

Section: Introductionmentioning

confidence: 99%

A two-level real-time vision machine combining coarse- and fine-grained parallelism

Jensen

Kjær-Nielsen

Pauwels

et al. 2010

J Real-Time Image Proc

Self Cite

View full text Add to dashboard Cite

In this paper, we describe a real-time vision machine having a stereo camera as input generating visual information on two different levels of abstraction. The system provides visual low-level and mid-level information in terms of dense stereo and optical flow, egomotion, indicating areas with independently moving objects as well as a condensed geometric description of the scene. The system operates at more than 20 Hz using a hybrid architecture consisting of one dual-GPU card and one quad-core CPU. The different processing stages of visual information have rather different characteristics that in some cases make fine-grained parallelization on a GPU less applicable. However, for most of the stages that are not efficiently implementable on a GPU, a coarse parallelization on multiple CPU-cores is applicable. We show that with such hybrid parallelism, we can achieve a speed up of approximately a factor 90 and a reduction of latency of a factor 26 compared to processing on a single CPU-core. Since the vision machine provides generic visual information it can be used in many contexts. Currently it is used in a driver assistance context as well as in two robotic applications.

show abstract

“…Based on this understanding, the ECOVISION consortium has developed a precursor of a real-time vision machine described in Wörgötter et al (2004). Furthermore, one application in the area of driver assistance systems has been developed as an industrial prototype (Mota et al, 2004).…”

Section: Editorial: Ecovision: Challenges In Early-cognitive Visionmentioning

confidence: 99%

Editorial: ECOVISION: Challenges in Early-Cognitive Vision

Krüger¹,

Wörgötter²,

Hulle³

2006

Int J Comput Vision

Self Cite

View full text Add to dashboard Cite

Early Cognitive Vision: Using Gestalt-Laws for Task-Dependent, Active Image-Processing

Cited by 18 publications

References 86 publications

Coding the presence of visual objects in a recurrent neural network of visual cortex

Coding the presence of visual objects in a recurrent neural network of visual cortex

A two-level real-time vision machine combining coarse- and fine-grained parallelism

Editorial: ECOVISION: Challenges in Early-Cognitive Vision

Contact Info

Product

Resources

About