BriefGPT.xyz
Jul, 2016
上下文问题:用递归神经网络提升视频中的物体检测
Context Matters: Refining Object Detection in Video with Recurrent Neural Networks
HTML
PDF
Subarna Tripathi, Zachary C. Lipton, Serge Belongie, Truong Nguyen
TL;DR
该论文提出了一种新的框架,通过捕捉时间空间和鼓励预测一致性来提高视频中的目标检测表现,并融合了强、弱监督的训练方式和平滑性惩罚,提高了 Youtube-Video Objects 数据集上的平均精度(mAP)。
Abstract
Given the vast amounts of
video
available online, and recent breakthroughs in
object detection
with static images,
object detection
in
→