Saliency-based input resampling for efficient object detection

• Main Authors: SHAMI, GODRATI, AMIT, POLIKLI, FARROKH, M, BENOUDI, BEHROUZ, AIT
• Format: Patent
• Language: Chinese; English
• Published: 18.06.2024
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

A method implemented using a processor for video processing includes receiving, via an artificial neural network (ANN), a video including a first frame and a second frame. A saliency map is generated based on the first frame of the video. The second frame of the video is sampled based on the saliency map. A first portion of the second frame is sampled at a first resolution and a second portion of the second frame is sampled at a second resolution. The first resolution is different from the second resolution. A resampled second frame is generated based on the sampling of the second frame. The resampled second frame is processed to determine an inference associated with the video. 一种使用进行视频处理的处理器实现的方法包括经由人工神经网络(ANN)接收包括第一帧和第二帧的视频。基于该视频的该第一帧来生成显著性图。基于该显著性图来对该视频的该第二帧进行采样。以第一分辨率对该第二帧的第一部分进行采样，并且以第二分辨率对该第二帧的第二部分进行采样。该第一分辨率不同于该第二分辨率。基于该第二帧的该采样来生成经重采样的第二帧。该经重采样的第二帧被处理以确定与该视频相关联的推断。