Attention and Prediction-Guided Motion Detection for Low-Contrast Small Moving Targets

Abstract

Small target motion detection within complex natural environments is an extremely challenging task for autonomous robots. Surprisingly, the visual systems of insects have evolved to be highly efficient in detecting mates and tracking prey, even though targets occupy as small as a few degrees of their visual fields. The excellent sensitivity to small target motion relies on a class of specialized neurons called small target motion detectors (STMDs). However, existing STMD-based models are heavily dependent on visual contrast and perform poorly in complex natural environments where small targets generally exhibit extremely low contrast against neighbouring backgrounds. In this paper, we develop an attention and prediction guided visual system to overcome this limitation. The developed visual system comprises three main subsystems, namely, an attention module, an STMD-based neural network, and a prediction module. The attention module searches for potential small targets in the predicted areas of the input image and enhances their contrast against complex background. The STMD-based neural network receives the contrast-enhanced image and discriminates small moving targets from background false positives. The prediction module foresees future positions of the detected targets and generates a prediction map for the attention module. The three subsystems are connected in a recurrent architecture allowing information to be processed sequentially to activate specific areas for small target detection. Extensive experiments on synthetic and real-world datasets demonstrate the effectiveness and superiority of the proposed visual system for detecting small, low-contrast moving targets against complex natural environments.

An Advanced Motion Prediction for Targeted Small Moving Objects

“…[6] An AGSS-VOS framework includes instance-agnostic, ainstancespecific and an attention-guided decoderfor single shot video-object segmentation. [7] An attentionpredictionguided visual system to overcome and hasthree subsystems, a) Acquisition Module b) Process Module c) Attention Module [8] A method to simulate the inference of human gaze behaviourincludes threestagesin 3D space. A coarsetofine strategy estimates a 3D gaze orientation from headrobustly.…”

EAPTON: Event-based Antinoise Powerlines Tracking with ON/OFF Enhancement

“…The ON and OFF channels in insects' Medulla are reported to play an essential role in the visual process for selecting attractive small targets while suppressing the response to irrelevant background noise [9], [10], [11]. However, it is still an open topic to determine how the ON/OFF mechanism enhances the neuron's preference for small targets.…”