Degraded Image Semantic Segmentation With Dense-Gram Networks

Guo, Dazhou; Pei, Yanting; Zheng, Kang; Yu, Hongkai; Lu, Yuhang; Wang, Song

doi:10.1109/tip.2019.2936111

Cited by 46 publications

(20 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…pretrained on clean images should be similar to the feature distribution of ω s (.) fine-tuned using degraded images (Guo et al 2019).…”

Section: Attention Transfer Network (Atn)mentioning

confidence: 99%

“…Similar to the discussion in (Guo et al 2019), to address such a problem, a straightforward approach is to use the degraded image, denoted as x d , as adversarial examples, together with the clear image, denoted as x o , to form the training dataset. However, obvious gaps exist between x o and x d in terms of semantics and details and simple adversarial training cannot effectively improve the performance (see the performance of adv-Amulet in the later experiments).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Type Self-Attention Guided Degraded Saliency Detection

Zhou

Wang

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

Existing saliency detection techniques are sensitive to image quality and perform poorly on degraded images. In this paper, we systematically analyze the current status of the research on detecting salient objects from degraded images and then propose a new multi-type self-attention network, namely MSANet, for degraded saliency detection. The main contributions include: 1) Applying attention transfer learning to promote semantic detail perception and internal feature mining of the target network on degraded images; 2) Developing a multi-type self-attention mechanism to achieve the weight recalculation of multi-scale features. By computing global and local attention scores, we obtain the weighted features of different scales, effectively suppress the interference of noise and redundant information, and achieve a more complete boundary extraction. The proposed MSANet converts low-quality inputs to high-quality saliency maps directly in an end-to-end fashion. Experiments on seven widely-used datasets show that our approach produces good performance on both clear and degraded images.

show abstract

“…pretrained on clean images should be similar to the feature distribution of ω s (.) fine-tuned using degraded images (Guo et al 2019).…”

Section: Attention Transfer Network (Atn)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Multi-Type Self-Attention Guided Degraded Saliency Detection

Zhou

Wang

et al. 2020

AAAI

Self Cite

View full text Add to dashboard Cite

show abstract

“…Li et al [29] proposed spatial-temporal regularized correlation filters (STRCF) which can be solved via the alternating direction method of multipliers. Inspired by the successful application of features from convolutional neural networks (CNN) in image classification [30], image segmentation [31] and image denoising [32], [33], Ma et al [34] utilized hierarchical convolutional features instead of handcrafted features in the framework of correlation filter to improve tracking performance. Qi et al [35] developed a novel adaptive weighted method to hedge each weak CNN based tracker into a stronger one.…”

Section: Related Work a Trackers Based On Correlation Filtermentioning

confidence: 99%

Robust Visual Object Tracking With Multiple Features and Reliable Re-Detection Scheme

Zhang

Guo

et al. 2020

IEEE Access

View full text Add to dashboard Cite

In recent years, correlation filter based trackers have seen widespread success because of their high efficiency and robustness. However, a single feature based tracker cannot deal with complex scenes such as serious occlusion, motion blur and illumination variation. In this paper, we develop a novel tracking method combining color feature, Hog feature and motion feature. The motion feature is estimated between adjacent frames by large displacement optical flow. Besides, in order to cope with boundary effect existing in traditional correlation filter based trackers, an adaptive cosine window is introduced in our method, which can highlight the target region, suppress the background region and enlarge search region. Meanwhile, a novel judge scheme combining Hog correlation response and color response is adopted to evaluate the reliability of tracking result. Finally, inverse sparse representation is presented to locate coarse positions of target in case of tracking failures. Extensive experiments on five famous tracking benchmarks including OTB100, TColor-128, UAVDT, UAV123 and VOT2016 demonstrate our proposed method outperform other sate-ofthe-art methods in terms of robustness and accuracy.

show abstract

“…Except for distances between predictions and ground truths directly measured, they can also be indirectly computed. For example, the reconstruction error between original labels and compressed ones by transformations applied in data‐dependent upsampling is minimised [33]; based on teacher–student networks, the dense gram matrix of feature maps in the source network should be close to that in the target model [20].…”

Section: Related Workmentioning

confidence: 99%

“…Some of them classify pixels using a single semantic segmentation model with pre‐trained weights [13–16, 18, 19, 21, 23, 25, 26, 29, 32, 33, 36]. Some other methods refine pixels using multiple convolutional neural networks [18, 20, 22, 24, 27, 28, 31, 34]. Still other methods post‐process pixels with additional modules based on low‐level boundary cues [30, 35].…”

Section: Introductionmentioning

confidence: 99%