Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.
Optimizing 3D Environment Reconstruction
Transcript of Optimizing 3D Environment Reconstruction
Yan (Asta) Li
Kang Zhang, Dr. Xin Li
A technique used to map unknown environment while identifying the pose (position and orientation) of the sensor
Implement and optimize simultaneous localization and mapping (SLAM); visualize accurate, dense 3D maps of unknown environments in real time
Microsoft Kinect: infrared (IR) depth sensor and color (RGB) camera
IR Depth Sensor
C++ compiled in MS Visual Studio;
Mobile Robot Programming Toolkit (MRPT) framework and OpenCV libraries
Mobile Robot Programming Toolkit (MRPT)
~ simple, minimal structure
~ familiar C++ source code
Scan unknown environment
Detect and track features using SIFT
Compute descriptors and correspondences
Register point clouds and estimate pose
Visualize global mapping
Scale-Invariant Feature Transform (SIFT): robust feature detection and tracking algorithm, invariant to scale, rotation, viewpoint, illumination, and image distortion.
Although the environment mapping is sparse, real-time and interactive frame rates during reconstruction are achieved with adequate accuracy.
~ Applications for unmanned aerial vehicles (UAVs)
~ Useful in GPS-denied environments
~ Assist with first-responders, police work, search and rescue
~ Real-time processing speed is a primary concern
~ Lack of graphics card stresses accuracy/time trade-offs
~ Hardware limitations: ineffective in direct sunlight, restricted maximum environment size
~ Accumulated position drift error (loop closure)