Video Saliency Detection by 3D Convolutional Neural Networks

Ding, Guanqun; Fang, Yuming

doi:10.1007/978-981-10-8108-8_23

Cited by 7 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Spatiotemporal features in video sequences are learned using a 3d convolutional network in [14] with three video frames as input of the network. A stack of convolutional layers similar to [57] followed by transposed convolutional layers are used in [64] to predict static saliency and then a similar network architecture is designed to predict dynamic saliency taking two consecutive video frames along with static saliency as prior.…”

Section: Video Saliency Detectionmentioning

confidence: 99%

“…Saliency detection has been an active research area in computer vision for a long time but most of the previous research has been focused on still-image saliency detection [27,49,11,53,36,41,10]. During the last couple of years, saliency detection in videos has gained a lot of interest as well [69,63,35,32,59,30,34,4,73,14,21,25,9,15,74,35,64]. Most of these methods try to incorporate motion cues into the previously designed saliency detection models and use them together with appearance features to predict video saliency.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Salient object detection in video using deep non-local neural networks

Shokri

Harati

Taba

2020

Journal of Visual Communication and Image Representation

View full text Add to dashboard Cite

Detection of salient objects in image and video is of great importance in many computer vision applications. In spite of the fact that the state of the art in saliency detection for still images has been changed substantially over the last few years, there have been few improvements in video saliency detection. This paper investigates the use of recently introduced non-local neural networks in video salient object detection. Non-local neural networks are applied to capture global dependencies and hence determine the salient objects. The effect of non-local operations is studied separately on static and dynamic saliency detection in order to exploit both appearance and motion features. A novel deep non-local neural network architecture is introduced for video salient object detection and tested on two well-known datasets DAVIS and FBMS. The experimental results show that the proposed algorithm outperforms state-of-the-art video saliency detection methods.

show abstract

Section: Video Saliency Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Salient object detection in video using deep non-local neural networks

Shokri

Harati

Taba

2020

Journal of Visual Communication and Image Representation

View full text Add to dashboard Cite

show abstract

“…[7], and the popularity of salient object detection approaches and data sets, e.g. [66,16,76]. Included in our set of reference models are two recent approaches: DeepVS (OMCNN-2CLSTM) [29] -code available via [30] -and ACLNet [75] -code available via [74].…”

Section: Reference Modelsmentioning

confidence: 99%

Supersaliency: A Novel Pipeline for Predicting Smooth Pursuit-Based Attention Improves Generalisability of Video Saliency

Startsev

Dörr

2020

IEEE Access

View full text Add to dashboard Cite

Predicting attention is a popular topic at the intersection of human and computer vision. However, even though most of the available video saliency data sets and models claim to target human observers' fixations, they fail to differentiate them from smooth pursuit (SP), a major eye movement type that is unique to perception of dynamic scenes. In this work, we highlight the importance of SP and its prediction (which we call supersaliency, due to greater selectivity compared to fixations), and aim to make its distinction from fixations explicit for computational models. To this end, we (i) use algorithmic and manual annotations of SP and fixations for two well-established video saliency data sets, (ii) train Slicing Convolutional Neural Networks for saliency prediction on either fixation-or SP-salient locations, and (iii) evaluate our and 26 publicly available dynamic saliency models on three data sets against traditional saliency and supersaliency ground truth. Overall, our models outperform the state of the art in both the new supersaliency and the traditional saliency problem settings, for which literature models are optimized. Importantly, on two independent data sets, our supersaliency model shows greater generalization ability and outperforms all other models, even for fixation prediction.

show abstract