Multispectral pedestrian detection: Benchmark dataset and baseline

By Soonmin Hwang, Jaesik Park, Namil Kim, Yukyung Choi, In-So Kweon. 2015

1,096 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Unsupervised Image-to-Image Translation Networks (2017, methodology): “In the first experiment, we trained a UNIT network using the KAIST multispectral pedestrian detection benchmark (Hwang et al., 2015) for translating between daytime and night time images and between thermal IR and RGB images.”
RFN-Nest: An end-to-end residual fusion network for infrared and visible images (2021, methodology): “For the second training stage, we choose the KAIST [43] dataset to train our RFN networks.”
RGB-T Object Tracking: Benchmark and Baseline (2018, background): “2) Due to the superior imaging mechanism, its alignment across modalities is more accurate, and does not require any pre-and post-processing ( e.g. , stereo matching [6, 7] and color correction [8]).”
DenseFuse: A Fusion Approach to Infrared and Visible Images (2018, neutral): “The input images are collected from [32].”
The Unmanned Aerial Vehicle Benchmark: Object Detection and Tracking (2018, neutral): “Caltech [14] 132k 347k D X 2012 KAIST [23] 95k 86k D X X 2015”
LLVIP: A Visible-infrared Paired Dataset for Low-light Vision (2021, methodology): “TNO [21], KAIST Multispectral Dataset [7], OTCBVS OSU Color-Thermal Database [3], etc.”
Fusion of Multispectral Data Through Illumination-aware Deep Neural Networks for Pedestrian Detection (2018, methodology): “With well-aligned visible and thermal image pairs with dense pedestrian annotations, the author proposed a new multispectral aggregated features (ACF+T+THOG) to process color-thermal image pairs and applied a boosted decision trees (BDT) for target classification.”
KAIST Multi-Spectral Day/Night Data Set for Autonomous and Assisted Driving (2018, methodology): “3) Multi-Spectral Pedestrian Dataset: Multi-spectral Pedestrian Dataset [13] is the first large-scale multi-spectral dataset for pedestrian detection in day and night.”
UNIT-DDPM: UNpaired Image Translation with Denoising Diffusion Probabilistic Models (2021, methodology): “Our method is evaluated against prior unpaired I2I translation methods [44] [25] [16] [24] on public datasets where ground truth input-output pairs are available [39] [44] [17].”
AdapNet: Adaptive semantic segmentation in adverse environmental conditions (2017, background): “In [9], the authors introduce a multi-spectral dataset containing RGB and thermal images for detecting pedestrians in non-ideal conditions.”
LasHeR: A Large-Scale High-Diversity Benchmark for RGBT Tracking (2021, background): “In the task of multi-modal object detection [10, 48, 49], multiple objects need to be localized and the global alignment should thus be required.”
LRRNet: A Novel Representation Learning Guided Fusion Network for Infrared and Visible Images (2023, background): “In recent years, several multi-modal datasets have been released: KAIST [25], RGBT-234 [26], and RoadScene [27].”
Improving Multispectral Pedestrian Detection by Addressing Modality Imbalance Problems (2020, methodology): “Comparisons with the state-of-the-art methods on the KAIST reasonable subset in terms of MR−2 [17] with different thresholds of IoU.”
Joint Monocular 3D Vehicle Detection and Tracking (2018, background): “KITTI-D [12] D C,P 7k - 41k X X KAIST [20] D P 50k - 42k X X KITTI-T [12] T C 8k 938 65k X X X MOT15-3D [24] T P 974 29 5632 X X MOT15-2D [24] T P 6k 500 40k X MOT16 [29] T C,P 5k 467 110k X X UA-DETRAC [45] D,T C 84k 6k 578k X X”
Multispectral Pedestrian Detection using Deep Fusion Convolutional Neural Networks (2016, methodology): “In [5], a feature-level fusion based approach is developed by extending the aggregated channel features detector of [6].”
Illumination-aware Faster R-CNN for Robust Multispectral Pedestrian Detection (2018, methodology): “We perform experiments under reasonable conﬁgura-tion [12].”
Multispectral Deep Neural Networks for Pedestrian Detection (2016, methodology): “All these models outperform the strong baseline detector Faster R-CNN on KAIST multispectral pedestrian dataset (KAIST) [17].”