Command Palette
Search for a command to run...
Endpoint Error EPE
End-Point Error (EPE) is an important evaluation metric for depth and binocular stereo estimation tasks in computer vision. It is the mean end-point error between predicted disparity/optical flow and ground-truth disparity/optical flow, measuring the absolute numerical accuracy of disparity maps/optical flow vectors pixel by pixel, and is commonly used for dense disparity and optical flow evaluation. The unit is pixel. This metric is widely used in relevant benchmarks and papers, and can effectively reflect model performance in this dimension. Lower values are better. It helps researchers and engineers quantify the actual performance differences of models in depth and binocular stereo estimation scenarios, facilitating horizontal comparison of different methods or model versions.
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.