强大的自我监督单眼深度估计用于内镜软组织变形场景与生物机械约束
概括
这项研究引入了一种新的自我监督学习方法,用于在内镜视频中准确的单眼深度估计. 它有效地处理镜面反射和软组织变形,提高了手术导航的准确性.
科学领域:
- 计算机视觉 计算机视觉
- 医疗成像医学成像
- 机器人技术 机器人技术 机器人技术
背景情况:
- 自主监督学习 (SSL) 在单眼深度和自我运动估计方面表现出色.
- 内镜场景带来了独特的挑战:镜面反射和软组织变形阻碍了SSL的准确性.
研究的目的:
- 开发一种强大的SSL方法,用于在具有挑战性的内镜环境中单眼深度估计.
- 通过解决镜面反射和软组织变形,提高准确性.
主要方法:
- 引入了一种图像距离转换策略,用于处理镜像反射.
- 提出了一个生物机械软组织变形约束来提高像素级准确性.
- 采用轻量级架构,以实现高效的计算和更快的推断.
主要成果:
- 在公共 (SCARED,SERV-CT) 和定制数据集上实现了强大的单眼深度估计.
- 证明了与最先进的 (SOTA) 方法相匹配的准确性和稳定性.
- 在SCARED数据集上,RMSE达到4.96mm,只有2.25M个参数.
结论:
- 拟议的方法有效地克服了内镜视觉中现有的SSL技术的局限性.
- 该方法显示了计算机辅助手术导航中临床应用的巨大潜力.
- 为实时性能提供精度,稳定性和计算效率的平衡.
相关概念视频
Muscles of the Eye
The muscles of the eye are sophisticated structures that control eye movement and focus, allowing for the precise and rapid adjustments necessary for vision. The human eye is controlled by ten muscles — six extraocular muscles, three intraocular muscles, and one primary eyelid retractor muscle.
Extraocular Muscles
The six extraocular muscles surround the eyeball and control its movements. They are responsible for a wide range of eye motions, including looking up, down, left, right, and rotating...
Extraocular Muscles
The six extraocular muscles surround the eyeball and control its movements. They are responsible for a wide range of eye motions, including looking up, down, left, right, and rotating...
Focusing of Light in the Eye
Light rays enter the eye through the cornea, a transparent dome-shaped tissue that is the eye's outermost layer. The cornea bends or refracts, light rays traveling to the pupil. The shape of the cornea determines how much of the light is bent and whether the image will be focused correctly on the retina at the back of the eye. Once the light has passed through both refraction layers, it converges into a single focal point onto a small area. This is where photoreceptors start transforming...


