YimianDai

2016-自动化学报-基于视觉的目标检测与跟踪综述

摘要

• 根据所处理对象的不同，将目标检测分为：
  • 基于背景建模的方法
  • 基于前景建模的方法
• 根据跟踪过程有无目标检测的参与，将跟踪方法分为：
  • 生成式
• 判别式

YimianDai

Notes on DCN

关键是要学到 transformation-invariant features，CNN 没有抽取 scale-invariant 特征的设计，因此也就没有 scale-invariant 性质

max-pooling [1] for small translation-invariance 因为我需要对位置很敏感，所以不应该有 max-pooling

除了 max-pooling for small translation-invariance 以外，CNN 中没有针对其他 geometric transformations 来 modeling

究其根本，是因为卷积操作本身具有固定的几何结构，而由其层叠搭建而成的卷积网络的几何结构也是固定的，所以不具有对于几何形变建模的能力。

YimianDai

Notes on Dilated Convolution

小物体信息无法重建 (假设有四个 pooling layer 则 任何小于 2^4 = 16 pixel 的物体信息将理论上无法重建。)

相比原来的正常 convolution，dilated convolution 多了一个 hyper-parameter 称之为 dilation rate 指的是 kernel 的间隔数量 (e.g. 正常的 convolution 是 dilatation rate 1)。

正常的是 dilation=(1, 1) kernel 的间隔数量，我觉得应该下 多少个 是另外一个 kernel 的点，正常的就是 下 1 个，当我

Reference

YimianDai

PR - 2013 - Sparse coding based visual tracking Review and experimental comparison

Abstract

• sparse coding methods in visual tracking can be categorized into
  • appearance modeling based on sparse coding (AMSC)
  • target searching based on sparse representation (TSSR)
  • their combination
• 结论
• (1) AMSC methods significantly outperform TSSR methods

YimianDai

2013 - KDD - Robust principal component analysis via capped norms

Abstract

• such convex formulation is based on a strong assumption which may not hold in real-world applications, and the approximation error in these convex relaxations often can- not be neglected.
  • 算是本文的 Motivation 吧，但是究竟原来的 convex relaxation 怎么在 real-world 里不 hold 了，也没有交代清楚
  • 总的来说，作者就是觉得大家都在用 convex relaxation 求解，他要用 non-convex 的来求解

Problem Formulation

YimianDai

2012 - TPAMI - Slic superpixels compared to state-of-the-art superpixel methods

SLIC SUPERPIXELS

Algorithm

• SLIC，唯一的一个参数就是 k，想要的 superpixel 的数量
• To produce roughly equally sized superpixels, the grid interval is $S= \sqrt{N/k}$
  • 最好当然是每个 superpixel 像素数都是一样的咯，是不是能够这么保证呢？拭目以待
• 怎么确定 center？

YimianDai

2012 - ECCV - Low-rank sparse learning for robust visual tracking

Abstract

• propose a new particle-filter based tracking algorithm that exploits the relationship between particles (candidate targets)
  • 就是 low-rank

Introduction

• visual tracking exploits the sparse representation of the target candidate using a dictionary of templates

YimianDai

NC - 2011 - Recent advances and trends in visual tracking A review

Introduction

对 tracking 算法的要求

• Robustness:
  • Robustness means that even under complicated conditions, the tracking algorithms should be able to follow the interested object. The tracking difficulties may be cluttered back- ground, partial and full changing illuminations, occlusions or complex object motion.
• Adaptivity:
• Additional to various changes of the environment that an object is located in, the object itself also undergoes changes. This requires a steady adaptation mechanism of the tracking system to the actual object appearance.

YimianDai

2009 - ICCV - Robust visual tracking using L1 minimization

摘要

• 一句话介绍本文方法：
  • a robust visual tracking method by casting tracking as a sparse approximation problem in a particle filter framework.
• 怎么处理 occlusion 和 corruption？
  • occlusion, corruption and other challenging issues are addressed seamlessly through a set of trivial templates.
• 怎么 detect the target？
• Specifically, to find the tracking target at a new frame, each target candidate is sparsely represented in the space spanned by target templates and trivial templates.

YimianDai

2005 - CVPR - Object recognition with features inspired by visual cortex

Introduction

Hierarchical approaches popular 的原因

• 和 hierarchical nature of primate visual cortex 相符合
• hierarchical approaches have been shown to consistently outperform flat single-template (holistic) object recognition systems
  • 就是效果好