ActiveZero++:混合域学习立体和基于信心的深度完成与零注释
IEEE transactions on pattern analysis and machine intelligence
|August 15, 2023
概括
ActiveZero++是主动立体视觉系统的新框架,它消除了对现实世界的深度数据的需求. 这种混合领域的学习方法实现了最先进的深度估计,超过商业传感器,缩小了Sim2Real的差距.
科学领域:
- 计算机视觉 计算机视觉
- 机器人技术 机器人技术 机器人技术
- 机器学习 机器学习
背景情况:
- 基于学习的立体声方法通常需要大量的真实世界的深度数据,这很难准确地获取.
- 模拟环境提供易于获得的地面真实深度,但弥合Sim2Real差距仍然是一个重大障碍.
研究的目的:
- 推出ActiveZero++,一种混合域学习框架,用于主动立体视觉,绕过了对现实世界的深度注释的需求.
- 在具有挑战性的地区使用新的自我监督技术来提高深度预测的准确性和稳定性.
主要方法:
- 在形状原始数据集上的模拟领域中使用监督和自我监督损失的组合.
- 在非分发的真实数据上采用自我监督的损失,引入时间红外 (IR) 重投损失以提高稳定性.
- 包含基于信任的深度完成模块,利用立体网络信任和深度正常一致性进行错误校正.
主要成果:
- 在真实世界数据上实现最先进的性能,在定性和定量评估中超过商业深度传感器.
- 显著减少了深度图的Sim2Real域差距,有利于下游任务,如6D姿势估计.
- 在深度预测中表现出更高的准确性和稳定性,特别是在难以感知的区域.
结论:
- ActiveZero++通过利用混合领域学习和新的自我监督损失,为主动立体视提供了一个有效的解决方案.
- 该框架成功地解决了现实世界的深度数据采集和Sim2Real域差距的局限性.
- 这种方法为推进现实应用中的深度估计提供了一个有希望的方向,而不需要大量的手动注释.
相关概念视频
01:15Depth Perception and Spatial Vision
720
Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.
720
01:18Uniform Depth Channel Flow: Problem Solving
87
To calculate the flow rate for a trapezoidal channel, first, identify the bottom width, side slope, and flow depth of the channel. The cross-sectional area (A) corresponding to the depth of flow (y), channel bottom width (B), and side slope (θ) is determined by:Next, calculate the wetted perimeter, which includes the bottom width and the sloped side lengths in contact with the water. Using the values of the cross-sectional area and the wetted perimeter, determine the hydraulic radius by...
87
01:27Uniform Depth Channel Flow
97
Uniform depth channel flow keeps fluid depth consistent along channels such as irrigation canals. In natural channels, such as rivers, approximate uniform flow is often assumed. This condition occurs when the channel’s bottom slope matches the energy slope, balancing potential energy lost from gravity with head loss due to shear stress. This balance prevents depth changes along the channel length, resulting in a steady, uniform flow.Uniform flow in open channels with a constant cross-section...
97
01:27Associative Learning
444
Associative learning is a fundamental concept in behavioral psychology, wherein a connection is established between two stimuli or events, leading to a learned response. This process is critical in understanding how behaviors are acquired and modified. Conditioning, the mechanism through which associations are formed, can be divided into two main types: classical conditioning and operant conditioning, each elucidating different aspects of associative learning.
Classical conditioning, also known...
Classical conditioning, also known...
444
01:12Differential Leveling
214
Differential leveling is a precise method in surveying used to determine the elevation difference between two points. Its primary goal is to establish accurate vertical measurements to create level surfaces or grade lines critical for designing and constructing infrastructures such as roads, bridges, and buildings.The procedure for differential leveling begins with setting up and leveling the instrument at a point where the benchmark can be seen. The level rod is held on the benchmark (BM), and...
214
01:26Introduction and Methods of Leveling
146
Leveling is a surveying procedure used to determine elevation differences between distant points. Elevation refers to the vertical distance above or below a reference datum, typically mean sea level (MSL). In the United States, elevations are often referenced to the mean sea level station at Father Point Rimouski along the St. Lawrence Seaway. To make the datum accessible, permanent markers are established throughout the region. These markers, called benchmarks, have known elevations. If the...
146