施工安全巡检易因视野遮挡导致特征缺失,出现危险误判情况。为提升施工风险识别效率,针对特征遮挡问题,以施工围栏检测为例,提出一种基于amodal补全技术的遮挡特征推理方法。首先,利用图像合成技术建立存在视野遮挡的围栏检测数据集。然...施工安全巡检易因视野遮挡导致特征缺失,出现危险误判情况。为提升施工风险识别效率,针对特征遮挡问题,以施工围栏检测为例,提出一种基于amodal补全技术的遮挡特征推理方法。首先,利用图像合成技术建立存在视野遮挡的围栏检测数据集。然后,结合YOLOv8实例分割和ASBU(amodal segmenter based on boundary uncertainty estimation)特征补全网络,推测围栏被遮挡部分的视觉特征。遮挡特征补全后的施工围栏可应用于闭合检测等多种施工安全监管任务。研究选择多个施工场地的围栏图像进行案例验证,可实现精准的围栏缺失特征补全(平均交并比mIoU>95.5%)。研究成果为施工场景的遮挡特征推理提供研究范式,有效提升了智能化施工安全监管效率。展开更多
Eyespots have been shown to perform an anti-predator function by intimidating predators.However,whether predators are deterred by the eyespots’mimicking eyes of higher-order predators,or by the conspicuousness of the...Eyespots have been shown to perform an anti-predator function by intimidating predators.However,whether predators are deterred by the eyespots’mimicking eyes of higher-order predators,or by the conspicuousness of the spots,is still being debated.The anti-predator mechanism of eyespots,which involves the predator’s visual perception and cognition,urges to be examined.To explore the“eye mimicry hypothesis”and“conspicuous signal hypothesis,”we designed prey with deceptive eyespots eliciting 2-dimensional(2D)or 3-dimensional(3D)visual illusions for domestic chicks(as predators).For 2D visual illusion,we provided prey with occluded eyespots for chicks.Chicks can“complete”the occluded eyespots into a full 2D concentric circle by the amodal completion mechanism,that is,one of visual illusion.For 3D visual illusion,we created the crescent-shaped sparkles on the eyespots to simulate a visually 3D eye.Our result indicated that(1)2D:chicks hesitated more when facing preys with occluded eyespots than facing preys with broken eyespots and(2)3D:compared to prey with no-sparkle eyespots,chicks spent more time approaching prey with sparkle eyespots and were more likely to approach them tangentially.Thus,the visual illusion perceived by chicks would impact the anti-predation effectiveness of eyespots.Eyespots that more closely resemble the real eyes,especially those with 3-dimensionality,provided more effective protection.Our study supported the eye mimicry hypothesis from the perspective of visual illusion.展开更多
Visual curve completion is a fundamental problem in understanding the principles of the human visual system. This problem is usually divided into two problems: a grouping problem and a shape problem.On one hand, thoug...Visual curve completion is a fundamental problem in understanding the principles of the human visual system. This problem is usually divided into two problems: a grouping problem and a shape problem.On one hand, though perception of the visually completed curve is clearly a global task(for example,a human perceives the Kanizsa triangle only when seeing all three black objects), conventional methods for solving the grouping problem are generally based on local Gestalt laws. On the other hand, the shape of the visually completed curve is usually recovered by minimizing shape energy in existing methods.However, not only do these methods lack mechanisms to adjust the shape of the recovered visual curve using perceptual, psychophysical, and neurophysiological knowledge, but it is also difficult to calculate an explicit representation of the visually completed curve. In this paper, we present a systematic computational model for generating a visually completed curve. Firstly, based on recent studies of perception, psychophysics, and neurophysiology, we formulate a grouping procedure based on the human visual system by seeking a minimum Hamiltonian cycle in a graph, solving the grouping problem in a global manner. Secondly, we employ a B′ezier curve-based model to represent the visually completed curve. Not only is an explicit representation deduced, but we also present a means to integrate knowledge from related areas, such as perception, psychophysics, and neurophysiology, and so on. The proposed computational model has been validated using many modal and amodal completion examples, and desirable results were obtained.展开更多
文摘施工安全巡检易因视野遮挡导致特征缺失,出现危险误判情况。为提升施工风险识别效率,针对特征遮挡问题,以施工围栏检测为例,提出一种基于amodal补全技术的遮挡特征推理方法。首先,利用图像合成技术建立存在视野遮挡的围栏检测数据集。然后,结合YOLOv8实例分割和ASBU(amodal segmenter based on boundary uncertainty estimation)特征补全网络,推测围栏被遮挡部分的视觉特征。遮挡特征补全后的施工围栏可应用于闭合检测等多种施工安全监管任务。研究选择多个施工场地的围栏图像进行案例验证,可实现精准的围栏缺失特征补全(平均交并比mIoU>95.5%)。研究成果为施工场景的遮挡特征推理提供研究范式,有效提升了智能化施工安全监管效率。
基金financially supported by the National Natural Science Foundation of China(Grant No.32471564)。
文摘Eyespots have been shown to perform an anti-predator function by intimidating predators.However,whether predators are deterred by the eyespots’mimicking eyes of higher-order predators,or by the conspicuousness of the spots,is still being debated.The anti-predator mechanism of eyespots,which involves the predator’s visual perception and cognition,urges to be examined.To explore the“eye mimicry hypothesis”and“conspicuous signal hypothesis,”we designed prey with deceptive eyespots eliciting 2-dimensional(2D)or 3-dimensional(3D)visual illusions for domestic chicks(as predators).For 2D visual illusion,we provided prey with occluded eyespots for chicks.Chicks can“complete”the occluded eyespots into a full 2D concentric circle by the amodal completion mechanism,that is,one of visual illusion.For 3D visual illusion,we created the crescent-shaped sparkles on the eyespots to simulate a visually 3D eye.Our result indicated that(1)2D:chicks hesitated more when facing preys with occluded eyespots than facing preys with broken eyespots and(2)3D:compared to prey with no-sparkle eyespots,chicks spent more time approaching prey with sparkle eyespots and were more likely to approach them tangentially.Thus,the visual illusion perceived by chicks would impact the anti-predation effectiveness of eyespots.Eyespots that more closely resemble the real eyes,especially those with 3-dimensionality,provided more effective protection.Our study supported the eye mimicry hypothesis from the perspective of visual illusion.
基金supported by the National Natural Science Foundation of China (Grant Nos. 61272300, 61379072, 61379069)the Key Technologies R&D Program of China (No. 2014BAK09B04)
文摘Visual curve completion is a fundamental problem in understanding the principles of the human visual system. This problem is usually divided into two problems: a grouping problem and a shape problem.On one hand, though perception of the visually completed curve is clearly a global task(for example,a human perceives the Kanizsa triangle only when seeing all three black objects), conventional methods for solving the grouping problem are generally based on local Gestalt laws. On the other hand, the shape of the visually completed curve is usually recovered by minimizing shape energy in existing methods.However, not only do these methods lack mechanisms to adjust the shape of the recovered visual curve using perceptual, psychophysical, and neurophysiological knowledge, but it is also difficult to calculate an explicit representation of the visually completed curve. In this paper, we present a systematic computational model for generating a visually completed curve. Firstly, based on recent studies of perception, psychophysics, and neurophysiology, we formulate a grouping procedure based on the human visual system by seeking a minimum Hamiltonian cycle in a graph, solving the grouping problem in a global manner. Secondly, we employ a B′ezier curve-based model to represent the visually completed curve. Not only is an explicit representation deduced, but we also present a means to integrate knowledge from related areas, such as perception, psychophysics, and neurophysiology, and so on. The proposed computational model has been validated using many modal and amodal completion examples, and desirable results were obtained.
