Visual Object Tracking via Multi-Stream Deep Similarity Learning Networks

Joint Correlation and Attention Based Feature Fusion Network for Accurate Visual Tracking

Correlation operation and attention mechanism are two popular feature fusion approaches which play an important role in visual object tracking. However, the correlation-based tracking networks are sensitive to location information but loss some context semantics, while the attention-based tracking networks can make full use of rich semantic information but ignore the position distribution of the tracked object. Therefore, in this paper, we propose a novel tracking framework based on joint correlation and attention networks, termed as JCAT, which can effectively combine the advantages of these two complementary feature fusion approaches. Concretely, the proposed JCAT approach adopts parallel correlation and attention branches to generate position and semantic features. Then the fusion features are obtained by directly adding the location feature and semantic feature. Finally, the fused features are fed into the segmentation network to generate the pixel-wise state estimation of the object. Furthermore, we develop a segmentation memory bank and an online sample filtering mechanism for robust segmentation and tracking. The extensive experimental results on eight challenging visual tracking benchmarks show that the proposed JCAT tracker achieves very promising tracking performance and sets a new state-of-the-art on the VOT2018 benchmark.

TDIOT: Target-Driven Inference for Deep Video Object Tracking

Recent tracking-by-detection approaches use deep object detectors as target detection baseline, because of their high performance on still images. For effective video object tracking, object detection is integrated with a data association step performed by either a custom design inference architecture or an end-to-end joint training for tracking purpose. In this work, we adopt the former approach and use the pre-trained Mask R-CNN deep object detector as the baseline. We introduce a novel inference architecture placed on top of FPN-ResNet101 backbone of Mask R-CNN to jointly perform detection and tracking, without requiring additional training for tracking purpose. The proposed single object tracker, TDIOT, applies an appearance similarity-based temporal matching for data association. To tackle tracking discontinuities, we incorporate a local search and matching module into the inference head layer that exploits SiamFC. Moreover, to improve robustness to scale changes, we introduce a scale adaptive region proposal network that enables to search for the target at an adaptively enlarged spatial neighborhood specified by the trace of the target. In order to meet long term tracking requirements, a low cost verification layer is incorporated into the inference architecture to monitor presence of the target based on its LBP histogram model. Performance evaluation on videos from VOT2016, VOT2018, and VOT-LT2018 datasets demonstrate that TDIOT achieves higher accuracy compared to the state-of-the-art short-term trackers while it provides comparable performance in long term tracking. We also compare our tracker on LaSOT dataset where we observe that TDIOT provides comparable performance with other methods that are trained on LaSOT. The source code and TDIOT output videos are accessible at https://github.com/msprITU/TDIOT.

Hierarchical Edge Refinement Network for Saliency Detection

At present, most saliency detection methods are based on fully convolutional neural networks (FCNs). However, FCNs usually blur the edges of salient objects. Due to that, the multiple convolution and pooling operations of the FCNs will limit the spatial resolution of the feature maps. To alleviate this issue and obtain accurate edges, we propose a hierarchical edge refinement network (HERNet) for accurate saliency detection. In detail, the HERNet is mainly composed of a saliency prediction network and an edge preserving network. Firstly, the saliency prediction network is used to roughly detect the regions of salient objects and is based on a modified U-Net structure. Then, the edge preserving network is used to accurately detect the edges of salient objects, and this network is mainly composed of the atrous spatial pyramid pooling (ASPP) module. Different from the previous indiscriminate supervision strategy, we adopt a new one-to-one hierarchical supervision strategy to supervise the different outputs of the entire network. Experimental results on five traditional benchmark datasets demonstrate that the proposed HERNet performs well when compared with the state-of-the-art methods.

Self-Supervised Deep Correlation Tracking

The training of a feature extraction network typically requires abundant manually annotated training samples, making this a time-consuming and costly process. Accordingly, we propose an effective self-supervised learning-based tracker in a deep correlation framework (named: self-SDCT). Motivated by the forward-backward tracking consistency of a robust tracker, we propose a multi-cycle consistency loss as self-supervised information for learning feature extraction network from adjacent video frames. At the training stage, we generate pseudo-labels of consecutive video frames by forward-backward prediction under a Siamese correlation tracking framework and utilize the proposed multi-cycle consistency loss to learn a feature extraction network. Furthermore, we propose a similarity dropout strategy to enable some low-quality training sample pairs to be dropped and also adopt a cycle trajectory consistency loss in each sample pair to improve the training loss function. At the tracking stage, we employ the pre-trained feature extraction network to extract features and utilize a Siamese correlation tracking framework to locate the target using forward tracking alone. Extensive experimental results indicate that the proposed self-supervised deep correlation tracker (self-SDCT) achieves competitive tracking performance contrasted to state-of-the-art supervised and unsupervised tracking methods on standard evaluation benchmarks.