Yi Cheng

The main challenge in video question answering (VideoQA) is to capture and understand the complex spatial and temporal relations between objects based on given questions. Existing graph-based methods for VideoQA usually ignore keywords in questions and employ a simple graph to aggregate features without considering relative relations between objects, which may lead to inferior performance. In this paper, we propose a Keyword-aware Relative Spatio-Temporal (KRST) graph network for VideoQA. First, to make question features aware of keywords, we employ an attention mechanism to assign high weights to keywords during question encoding. The keyword-aware question features are then used to guide video graph construction. Second, because relations are relative, we integrate the relative relation modeling to better capture the spatio-temporal dynamics among object nodes. Moreover, we disentangle the spatio-temporal reasoning into an object-level spatial graph and a frame-level temporal graph, which reduces the impact of spatial and temporal relation reasoning on each other. Extensive experiments on the TGIF-QA, MSVD-QA and MSRVTT-QA datasets demonstrate the superiority of our KRST over multiple state-of-the-art methods.

Although object detection has achieved significant progress in the past decade, detecting small objects is still far from satisfactory due to the high variability of object scales and complex backgrounds. The common way to enhance small object detection is to use high-resolution (HR) images. However, this method incurs huge computational resources which grow squarely with the resolution of images. To achieve both accuracy and efficiency, we propose a novel reinforcement learning framework that employs an efficient policy network consisting of a Spatial Transformation Network to enhance the state representation learning and a Transformer model with early convolution to improve feature extraction. Our method has two main steps: (1) coarse location query (CLQ), where an RL agent is trained to predict the locations of small objects on low-resolution (LR) (down-sampled version of HR) images; (2) context-sensitive object detection where HR image patches are used to detect objects on the selected coarse locations and LR image patches on background areas (containing no small objects). In this way, we can obtain high detection performance on small objects while avoiding unnecessary computation on background areas. The proposed method has been tested and benchmarked on various datasets. On the Caltech Pedestrians Detection and Web Pedestrians datasets, the proposed method improves the detection accuracy by 2%, while reducing the number of processed pixels. On the Vision meets Drone object detection dataset and the Oil and Gas Storage Tank dataset, the proposed method outperforms the state-of-the-art (SotA) methods. On MS COCO mini-val set, our method outperforms SotA methods on small object detection, while also achieving comparable performance on medium and large objects.

View-based methods have achieved state-of-the-art performance in 3D object retrieval. However, view-based methods still encounter two major challenges. The first is how to leverage the inter-view correlation to enhance view-level visual features. The second is how to effectively fuse view-level features into a discriminative global descriptor. Towards these two challenges, we propose a multi-range view aggregation network (MRVANet) with a vision transformer based feature fusion scheme for 3D object retrieval. Unlike the existing methods which only consider aggregating neighboring or adjacent views which could bring in redundant information, we propose a multi-range view aggregation module to enhance individual view representations through view aggregation beyond only neighboring views but also incorporate the views at different ranges. Furthermore, to generate the global descriptor from view-level features, we propose to employ the multi-head self-attention mechanism introduced by vision transformer to fuse the view-level features. Extensive experiments conducted on three public datasets including ModelNet40, ShapeNet Core55 and MCB-A demonstrate the superiority of the proposed network over the state-of-the-art methods in 3D object retrieval.

Models for image semantics understanding, such as deep learning (DL) models and mathematical models, are often trained on specific dataset or configured with specific parameters. When deploying such models on new tasks in a different test environment, it requires considerable effort to re-train the model or extensive expertise to tune the parameters. In this paper, we propose a smart reinforcement learning (RL) agent that could learn to tune parameters automatically to enhance model performance. The learning process is formulated as a generic control task for parameter adjustment, and applied to two use scenarios: (1) image attributes tuning to improve object detection performance on fixed DL model, and (2) parameter tuning of the mathematical model (Level Set) for image segmentation. We design a novel dynamic threshold mechanism in a multi-branch RL agent to effectively tune parameters of image qualities (for object detection) and Level Set models (for object segmentation). We conduct experiments on Pascal-VOC testing set, MS COCO validation set and a proprietary dataset of industrial components, where we achieve substantial improvement on object detection accuracy. We also perform experiments on the automatic parameter tuning of Level Set models. Results show that our method facilitates considerable performance improvement on public datasets compared with baseline method.