Command Palette
Search for a command to run...
Threshold Accuracy δ<1.25
Threshold Accuracy (δ < 1.25) is an important evaluation metric for depth and stereo estimation tasks in computer vision. It is the proportion of pixels whose ratio of predicted depth to ground-truth depth falls within the interval [1/1.25, 1.25], measuring the accuracy of depth estimation within the threshold tolerance. Multi-threshold combinations such as δ<1.25 and δ<1.25² can comprehensively characterize performance. It is usually reported as a percentage or a 0–1 score. This metric is widely used in related benchmarks and papers and can effectively reflect the model's performance in this dimension. The higher the value, the better.
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.