Annotating a large-scale in-the-wild person re-identification dataset especially of marathon runners is a challenging task. The variations in the scenarios such as camera viewpoints, resolution, occlusion, and illumination make the problem non-trivial. Manually annotating bounding boxes in such large-scale datasets is cost-inefficient. Additionally, due to crowdedness and occlusion in the videos, aligning the identity of runners across multiple disjoint cameras is a challenge. We collected a novel large-scale in-the-wild video dataset of marathon runners. The dataset consists of hours of recording of thousands of runners captured using 42 hand-held smartphone cameras and covering real-world scenarios. Due to the presence of crowdedness and occlusion in the videos, the annotation of runners becomes a challenging task. We propose a new scheme for tackling the challenges in the annotation of such large dataset. Our technique reduces the overall cost of annotation in terms of time as well as budget. We demonstrate performing fps analysis to reduce the effort and time of annotation. We investigate several annotation methods for efficiently generating tight bounding boxes. Our results prove that interpolating bounding boxes between keyframes is the most efficient method of bounding box generation amongst several other methods and is 3x times faster than the naive baseline method. We introduce a novel way of aligning the identity of runners in disjoint cameras. Our inter-camera alignment tool integrated with the state-of-the-art person re-id system proves to be sufficient and effective in the alignment of the runners across multiple cameras with non-overlapping views. Our proposed framework of annotation reduces the annotation cost of the dataset by a factor of 16x, also effectively aligning 93.64\% of the runners in the cross-camera setting.

Person re-identification based on appearance is challenging due to varying views and lighting conditions in different cameras, or when multiple persons wear similar clothing styles and color. Considering these challenges, gait patterns provide an alternative to appearance, as gait can be captured from a distance and at a low resolution. In this paper we investigate and evaluate running gait as a unique attribute for video person re-identification in a recreational long-distance running event with 257 participants. We show that running gait recognition achieves competitive performance compared to video-based approaches in the cross-camera retrieval task and that gait and appearance features are complementary to each other. In addition, we compare gait recognition applied to walking and running sequences. An important difference is that we walk with straight arms, but run with bent arms. We propose to use human semantic parsing to create partial gait silhouettes from body parts to find the most discriminative combination. We demonstrate that the arm and leg swing are the most discriminative parts of the running gait. Our proposed method provides better recognition results by removing the torso from the silhouettes and allowing the arm swing to be more visible.

Deep learning has significantly improved Re-Id per- formance but it requires a large amount of data, however, obtaining data is expensive from both time and money perspective. Inspired by ImageNet pre- trained models and synthetic data generation techniques, this paper investigates to utilise real-world and syn- thetic Re-Id datasets to augment task performance. Firstly, we propose two methods to apply external Re-Id data, NDTL (Neighbour-Domain Transfer Learning) and NDDS (Neighbour-Domain Data Stitching). Secondly, we quantitatively illustrate that both real-world and syn- thetic data could mitigate Re-Id data shortage problems, using Re-Id dataset to pre-train models is better than us- ing ImageNet, we achieve up to 28.2% mAP improvement on DukeMTMC and 5.2% on Market-1501. Moreover, we find out that viewpoint, one of Re-Id relevant factors, has the an influence on the system performance due to viewpoint-wise non-alignment and unbalance of the orig- inal dataset, it also assists the performance if train set is augmented balanced. Our research strongly illustrates both real-world and synthetic Re-Id dataset can effec- tively augment Re-Id task, viewpoint is an essential fac- tor and based on which, train-test distribution dramati- cally influences Re-Id performance, and balancing train classes are also helpful to improve the performance.

Visualizing runners trajectory from video data is not straightforward because the video data does not contain the explicit information of which runners appear in the video. Only the visual information related to the runner, such as runner’s unique ID (called bib number), is available. To this end, we propose two automatic runner detection methods, i.e. scene text detection which identifies the runners by detecting their bib number and person re-identification which detects the runners based on their appearance. To evaluate the proposed methods, we create a ground truth database from the video dataset, which consists of video and frame interval information where the runners appear. The video dataset was recorded by nine cameras at different locations during the Campus Run 2018 event. The experimental evidence shows that the scene text recognition method achieves up to 74.05 for F1-score and person re identification achieves up to 87.76 for F1-score. To conclude, we find that the person re-identification method outperforms the scene text recognition method.