Depth Estimation
A computer vision task that predicts the depth distance of each pixel in a scene from image data. It is categorized into relative depth estimation, which infers ordinal depth ordering, and metric absolute depth estimation, which recovers true physical distances.
Timeline (last 90 days)
One-way pipeline methods are prone to depth estimation failures on transparent surfaces or weakly textured regions, leading to problems in the final scene such as misaligned walls, floating objects, and distorted proportions.
Highly specular objects can cause depth sensors to misinterpret mirror images as real space, and object detection models may be misled by reflected virtual images
Layer separation technology integrates three core modules: semantic segmentation, depth estimation, and motion prediction.
All Facts (3)
One-way pipeline methods are prone to depth estimation failures on transparent surfaces or weakly textured regions, leading to problems in the final scene such as misaligned walls, floating objects, and distorted proportions.
50%UnverifiedHighly specular objects can cause depth sensors to misinterpret mirror images as real space, and object detection models may be misled by reflected virtual images
50%UnverifiedLayer separation technology integrates three core modules: semantic segmentation, depth estimation, and motion prediction.
50%