scholarly journals Bottom-up and Layerwise Domain Adaptation for Pedestrian Detection in Thermal Images

Author(s):  
My Kieu ◽  
Andrew D. Bagdanov ◽  
Marco Bertini

Pedestrian detection is a canonical problem for safety and security applications, and it remains a challenging problem due to the highly variable lighting conditions in which pedestrians must be detected. This article investigates several domain adaptation approaches to adapt RGB-trained detectors to the thermal domain. Building on our earlier work on domain adaptation for privacy-preserving pedestrian detection, we conducted an extensive experimental evaluation comparing top-down and bottom-up domain adaptation and also propose two new bottom-up domain adaptation strategies. For top-down domain adaptation, we leverage a detector pre-trained on RGB imagery and efficiently adapt it to perform pedestrian detection in the thermal domain. Our bottom-up domain adaptation approaches include two steps: first, training an adapter segment corresponding to initial layers of the RGB-trained detector adapts to the new input distribution; then, we reconnect the adapter segment to the original RGB-trained detector for final adaptation with a top-down loss. To the best of our knowledge, our bottom-up domain adaptation approaches outperform the best-performing single-modality pedestrian detection results on KAIST and outperform the state of the art on FLIR.

Author(s):  
Lianli Gao ◽  
Zhilong Zhou ◽  
Heng Tao Shen ◽  
Jingkuan Song

Image edge detection is considered as a cornerstone task in computer vision. Due to the nature of hierarchical representations learned in CNN, it is intuitive to design side networks utilizing the richer convolutional features to improve the edge detection. However, there is no consensus way to integrate the hierarchical information. In this paper, we propose an effective and end-to-end framework, named Bidirectional Additive Net (BAN), for image edge detection. In the proposed framework, we focus on two main problems: 1) how to design a universal network for incorporating hierarchical information sufficiently; and 2) how to achieve effective information flow between different stages and gradually improve the edge map stage by stage. To tackle these problems, we design a consecutive bottom-up and top-down architecture, where a bottom-up branch can gradually remove detailed or sharp boundaries to enable accurate edge detection and a top-down branch offers a chance of error-correcting by revisiting the low-level features that contain rich textual and spatial information. And attended additive module (AAM) is designed to cumulatively refine edges by selecting pivotal features in each stage. Experimental results show that our proposed methods can improve the edge detection performance to new records and achieve state-of-the-art results on two public benchmarks: BSDS500 and NYUDv2.


Author(s):  
Céline Hocquette ◽  
Stephen H. Muggleton

Predicate Invention in Meta-Interpretive Learning (MIL) is generally based on a top-down approach, and the search for a consistent hypothesis is carried out starting from the positive examples as goals. We consider augmenting top-down MIL systems with a bottom-up step during which the background knowledge is generalised with an extension of the immediate consequence operator for second-order logic programs. This new method provides a way to perform extensive predicate invention useful for feature discovery. We demonstrate this method is complete with respect to a fragment of dyadic datalog. We theoretically prove this method reduces the number of clauses to be learned for the top-down learner, which in turn can reduce the sample complexity. We formalise an equivalence relation for predicates which is used to eliminate redundant predicates. Our experimental results suggest pairing the state-of-the-art MIL system Metagol with an initial bottom-up step can significantly improve learning performance.


2020 ◽  
Vol 34 (07) ◽  
pp. 10551-10558 ◽  
Author(s):  
Long Chen ◽  
Chujie Lu ◽  
Siliang Tang ◽  
Jun Xiao ◽  
Dong Zhang ◽  
...  

In this paper, we focus on the task query-based video localization, i.e., localizing a query in a long and untrimmed video. The prevailing solutions for this problem can be grouped into two categories: i) Top-down approach: It pre-cuts the video into a set of moment candidates, then it does classification and regression for each candidate; ii) Bottom-up approach: It injects the whole query content into each video frame, then it predicts the probabilities of each frame as a ground truth segment boundary (i.e., start or end). Both two frameworks have respective shortcomings: the top-down models suffer from heavy computations and they are sensitive to the heuristic rules, while the performance of bottom-up models is behind the performance of top-down counterpart thus far. However, we argue that the performance of bottom-up framework is severely underestimated by current unreasonable designs, including both the backbone and head network. To this end, we design a novel bottom-up model: Graph-FPN with Dense Predictions (GDP). For the backbone, GDP firstly generates a frame feature pyramid to capture multi-level semantics, then it utilizes graph convolution to encode the plentiful scene relationships, which incidentally mitigates the semantic gaps in the multi-scale feature pyramid. For the head network, GDP regards all frames falling in the ground truth segment as the foreground, and each foreground frame regresses the unique distances from its location to bi-directional boundaries. Extensive experiments on two challenging query-based video localization tasks (natural language video localization and video relocalization), involving four challenging benchmarks (TACoS, Charades-STA, ActivityNet Captions, and Activity-VRL), have shown that GDP surpasses the state-of-the-art top-down models.


Author(s):  
Xiao Wang ◽  
Jun Chen ◽  
Zheng Wang ◽  
Wu Liu ◽  
Shin'ichi Satoh ◽  
...  

Pedestrian detection at nighttime is a crucial and frontier problem in surveillance, but has not been well explored by the computer vision and artificial intelligence communities. Most of existing methods detect pedestrians under favorable lighting conditions (e.g. daytime) and achieve promising performances. In contrast, they often fail under unstable lighting conditions (e.g. nighttime). Night is a critical time for criminal suspects to act in the field of security. The existing nighttime pedestrian detection dataset is captured by a car camera, specially designed for autonomous driving scenarios. The dataset for nighttime surveillance scenario is still vacant. There are vast differences between autonomous driving and surveillance, including viewpoint and illumination. In this paper, we build a novel pedestrian detection dataset from the nighttime surveillance aspect: NightSurveillance1. As a benchmark dataset for pedestrian detection at nighttime, we compare the performances of state-of-the-art pedestrian detectors and the results reveal that the methods cannot solve all the challenging problems of NightSurveillance. We believe that NightSurveillance can further advance the research of pedestrian detection, especially in the field of surveillance security at nighttime.


2014 ◽  
Vol 2014 ◽  
pp. 1-7
Author(s):  
Yu Li-ping ◽  
Tang Huan-ling ◽  
An Zhi-yong

Pedestrian detection is an active area of research in computer vision. It remains a quite challenging problem in many applications where many factors cause a mismatch between source dataset used to train the pedestrian detector and samples in the target scene. In this paper, we propose a novel domain adaptation model for merging plentiful source domain samples with scared target domain samples to create a scene-specific pedestrian detector that performs as well as rich target domain simples are present. Our approach combines the boosting-based learning algorithm with an entropy-based transferability, which is derived from the prediction consistency with the source classifications, to selectively choose the samples showing positive transferability in source domains to the target domain. Experimental results show that our approach can improve the detection rate, especially with the insufficient labeled data in target scene.


2020 ◽  
Vol 49 (43) ◽  
pp. 15139-15148
Author(s):  
César L. Ruiz-Zambrana ◽  
Magdalena Malankowska ◽  
Joaquín Coronas

This perspective comprehensively summarizes the recent state of the art in the use of top-down and bottom-up methodologies to create metal organic framework (MOF) structures with a defined pattern at the nano- and micro-scale.


Sensors ◽  
2021 ◽  
Vol 21 (15) ◽  
pp. 5153
Author(s):  
Marcel Tintelott ◽  
Vivek Pachauri ◽  
Sven Ingebrandt ◽  
Xuan Thang Vu

Silicon nanowire field-effect transistors (SiNW-FET) have been studied as ultra-high sensitive sensors for the detection of biomolecules, metal ions, gas molecules and as an interface for biological systems due to their remarkable electronic properties. “Bottom-up” or “top-down” approaches that are used for the fabrication of SiNW-FET sensors have their respective limitations in terms of technology development. The “bottom-up” approach allows the synthesis of silicon nanowires (SiNW) in the range from a few nm to hundreds of nm in diameter. However, it is technologically challenging to realize reproducible bottom-up devices on a large scale for clinical biosensing applications. The top-down approach involves state-of-the-art lithography and nanofabrication techniques to cast SiNW down to a few 10s of nanometers in diameter out of high-quality Silicon-on-Insulator (SOI) wafers in a controlled environment, enabling the large-scale fabrication of sensors for a myriad of applications. The possibility of their wafer-scale integration in standard semiconductor processes makes SiNW-FETs one of the most promising candidates for the next generation of biosensor platforms for applications in healthcare and medicine. Although advanced fabrication techniques are employed for fabricating SiNW, the sensor-to-sensor variation in the fabrication processes is one of the limiting factors for a large-scale production towards commercial applications. To provide a detailed overview of the technical aspects responsible for this sensor-to-sensor variation, we critically review and discuss the fundamental aspects that could lead to such a sensor-to-sensor variation, focusing on fabrication parameters and processes described in the state-of-the-art literature. Furthermore, we discuss the impact of functionalization aspects, surface modification, and system integration of the SiNW-FET biosensors on post-fabrication-induced sensor-to-sensor variations for biosensing experiments.


Author(s):  
Yibing Song ◽  
Jiawei Zhang ◽  
Linchao Bao ◽  
Qingxiong Yang

Exemplar-based face sketch synthesis methods usually meet the challenging problem that input photos are captured in different lighting conditions from training photos. The critical step causing the failure is the search of similar patch candidates for an input photo patch. Conventional illumination invariant patch distances are adopted rather than directly relying on pixel intensity difference, but they will fail when local contrast within a patch changes. In this paper, we propose a fast preprocessing method named Bidirectional Luminance Remapping (BLR), which interactively adjust the lighting of training and input photos. Our method can be directly integrated into state-of-the-art exemplar-based methods to improve their robustness with ignorable computational cost


Author(s):  
Tanja A. Börzel ◽  
Diana Panke

This chapter examines the concept of Europeanization and why it has become prominent in research on the European Union and its member states. It first explains what Europeanization means before discussing the main approaches used in studying Europeanization. It then reviews the state of the art with particular reference to ‘top-down’ Europeanization (how the EU affects states) and illustrates the theoretical arguments with empirical examples. It also considers ‘bottom-up’ Europeanization (how states can influence the EU), offers some theoretical explanations for the empirical patterns observed, and provides an overview of research that explores the relationship between bottom-up and top-down Europeanization. Finally, it reflects on the future of Europeanization research and suggests that Europeanization will continue to be an important field of EU research for the foreseeable future.


PsycCRITIQUES ◽  
2005 ◽  
Vol 50 (19) ◽  
Author(s):  
Michael Cole
Keyword(s):  
Top Down ◽  

Sign in / Sign up

Export Citation Format

Share Document