> ## Documentation Index
> Fetch the complete documentation index at: https://docs.uplift.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Tips for high-quality motion capture

> Ensure your captures are of the highest quality and accuracy by following these steps.

Uplift's 2 camera capture provides a reliable method of motion capture that can be easily set up at any location to capture a number of sports movements. Because of the portable and flexible nature of Uplift's solution, the environment, positioning, lighting, and other variables can change from capture to capture. To make sure you're always capturing the highest-quality movements no matter the situation, follow these steps.

# Use solid tripods and mounts

Solid tripods and mounts ensure solid video. See our [recommended capture equipment](../getting-started/recommended-capture-equipment.md).

# Make sure the cameras are 90 degrees offset from each other, and measure your distances

Our 2-camera capture model works by reconstructing 3D movement assuming the cameras are 90 degrees offset from each other; if they aren't, the resulting metrics and recommendations could be affected. Make sure you're following the camera positioning guides in-app.

Also, measure the distance between the camera and the athlete on each device! While it may be tempting to guestimate, our models use those distance to calculate displacement metrics like stride length, drifting foreward, etc.

[This article](camera-setup-2-camera-assessments.md) include setup guidance and recommnedations on how to measure once, note the distance, and then mark the ground so you never have to measure again.

# Make sure the athlete's whole body is in frame for the entire movement

If we can't see something in the video, we can't track it. Make sure the athlete does not start or end thair movement out-of-frame in the video window. Doing a test movement after setting up the devices can help.

# Use good lighting

The model relies on the contrast in the video to detect the joint centers of the body. *(For an example of how these algorithms work[, you can check out this article outlining how these algorithms may learn to recognize a bird](https://machinelearningmastery.com/how-to-visualize-filters-and-feature-maps-in-convolutional-neural-networks/))*. That means **if the video is too dark**, or if **the athlete is back-lit so they look like a silhouette**, our models won't be able to track their joints, and thus the captures may fail QA.

# Avoid baggy clothing and dark colors if possible

Uplift is working to detect your joint centers, so for the ssme reason poor lighting can cause issues, baggy clothing will interfere with the processing. The algorithm works by looking at edges, contours, and colors in the image. Dark colors may not show as much contrast in the recording. Excessively baggy clothing will throw off the edge detection in the model and cause it to misread your athlete's joint centers and body position.

# Perform one movement per video recording

Make sure to only perform a single jump, swing, pitch, etc. for each video clip you record. Multiple movements in a single clip will cause our event detection to fail.

# Ensure other people aren’t in the video recording

When other people are in the video recording, it’s possible that the model gets confused when it sees keypoints on other individuals. If the athlete you're assessing is in the foreground and others are way in the background, it's usually not an issue, but if there are others close to the athlete in the video frame, it can cause issues.