Tracking the Saliency Features in Images Based on Human Observation Statistics

Szalai, Szilard; Tâm,; Szir, s; nyi,; Vidnyánszky, Zoltán

doi:10.1007/978-3-642-32436-9_19

Cited by 1 publication

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, our method with randomly weighted graphs is evaluated as a reference. More about the verification can be found in [13].…”

Section: Comparison Of the Estimation To Human Ground Truthmentioning

confidence: 99%

“…Our experiments characterize human eye movement with intentional factors. The experimental setup and conditions are described in details in our previous work [13].…”

Section: Training Our Model a Human Perceptional Experimentsmentioning

confidence: 99%

“…This fusion is the basis of our saliency model. More details are presented in our previous work [13].…”

Section: Fusing Human Perception and Machine Visionmentioning

confidence: 99%

“…Where eye movement and top-down factors are investigated, providing the presented circumstances is essential. More about the evaluation test can be found in [13]. …”

Section: A Psychophysical Evaluation Experimentsmentioning

confidence: 99%

“…Moreover, this method is comparable with more sophisticated bottom-up methods. We objectively compared our method with [1] and [2] to human ground truth data by the Kullback-Leibler divergence (Table I) as in [13], however; in this paper the robustness of our approach is investigated. Our method gives better saliency estimation than other approaches; the role of human training is proven by comparing to randomly weighted graphs.…”

Section: Numerical Evaluation Of the Modelmentioning

confidence: 99%

See 4 more Smart Citations

Saliency map estimation by constructing graphs of possible eye-tracking paths

Szalai

Szirányi

Vidnyánszky

2012

2012 IEEE 3rd International Conference on Cognitive Infocommunications (CogInfoCom)

Self Cite

View full text Add to dashboard Cite

We introduce a novel approach for saliency detection where we fuse perceptional saliency with machine saliency in a statistical approach. The improvement of our fused algorithm against other methods is presented. Human saliency is recorded from human eye movement during free view training. The transition movements caused by the saccades are evaluated to generate transition probability tables. This new kind of training is applied into a connection graph based model where transitions among the machine generated saliency points are weighted by the probabilities derived from the human trained probability table. In the presented method, different psychophysical studies are taken into consideration by inferring regions of interests. The proposed method results in a good estimation of the possible interest areas of the human vision measurements.

show abstract

“…Moreover, our method with randomly weighted graphs is evaluated as a reference. More about the verification can be found in [13].…”

Section: Comparison Of the Estimation To Human Ground Truthmentioning

confidence: 99%

“…Our experiments characterize human eye movement with intentional factors. The experimental setup and conditions are described in details in our previous work [13].…”

Section: Training Our Model a Human Perceptional Experimentsmentioning

confidence: 99%

“…This fusion is the basis of our saliency model. More details are presented in our previous work [13].…”

Section: Fusing Human Perception and Machine Visionmentioning

confidence: 99%

“…Where eye movement and top-down factors are investigated, providing the presented circumstances is essential. More about the evaluation test can be found in [13]. …”

Section: A Psychophysical Evaluation Experimentsmentioning

confidence: 99%

Section: Numerical Evaluation Of the Modelmentioning

confidence: 99%

See 3 more Smart Citations