Solving jigsaw puzzles has been extensively studied. While most existing models focus on solving either small-scale puzzles or puzzles with no gap between fragments, solving large-scale puzzles with gaps presents distinctive challenges in both image understanding and combinatorial optimization. To tackle these challenges, we propose a framework of Evolutionary Reinforcement Learning with Multi-head Puzzle Perception (ERL-MPP) to derive a better set of swapping actions for solving the puzzles. Specifically, to tackle the challenges of perceiving the puzzle with gaps, a Multi-head Puzzle Perception Network (MPPN) with a shared encoder is designed, where multiple puzzlet heads comprehensively perceive the local assembly status, and a discriminator head provides a global assessment of the puzzle. To explore the large swapping action space efficiently, an Evolutionary Reinforcement Learning (EvoRL) agent is designed, where an actor recommends a set of suitable swapping actions from a large action space based on the perceived puzzle status, a critic updates the actor using the estimated rewards and the puzzle status, and an evaluator coupled with evolutionary strategies evolves the actions aligning with the historical assembly experience. The proposed ERL-MPP is comprehensively evaluated on the JPLEG-5 dataset with large gaps and the MIT dataset with large-scale puzzles. It significantly outperforms all state-of-the-art models on both datasets.

Acquiring commonsense knowledge about entity-pairs from images is crucial across diverse applications. Distantly supervised learning has made significant advancements by automatically retrieving images containing entity pairs and summarizing commonsense knowledge from the bag of images. However, the retrieved images may not always cover all possible relations, and the informative features across the bag of images are often overlooked. To address these challenges, a Multi-modal Cross-domain Feature Learning framework is proposed to incorporate the general domain knowledge from a large vision-text foundation model, ViT-GPT2, to handle unseen relations and exploit complementary information from multiple sources. Then, a Group Attention module is designed to exploit the attentive information from other instances of the same bag to boost the informative features of individual instances. Finally, a Gamma-corrected Gated Fusion is designed to select a subset of informative instances for a comprehensive summarization of commonsense entity relations. Extensive experimental results demonstrate the superiority of the proposed method over state-of-the-art models for extracting commonsense knowledge.

Although 2D radar signal representations, such as spectrograms and range-Doppler maps have been widely used for target recognition, 3D time-range-Doppler (TRD) has been less studied, partially because of the difficulties in extracting features from the TRD representation, i.e., shallow 3D neural networks have limited discriminant power, but repeatedly applying 3D convolutions will lead to an oversized 3D network. A hybrid 3D–2D network architecture, Progressively-Orthogonally-Mapped EfficientNet (POMEN), is proposed to address these challenges. More specifically, the proposed POMEN utilizes 3D convolutions in the earlier stages to capture the information embedded in the sparse 3D TRD representation, and to avoid the oversized feature map caused by excessively applying 3D convolutions, we propose to progressively map the 3D features into three sets of 2D features corresponding to the range-time signature, range-Doppler map and time-Doppler signature (spectrogram), respectively. Subsequently, 2D EfficientNet blocks were designed to extract discriminant information from the three sets of 2D feature maps. This hybrid 3D–2D network design effectively extracts features from the 3D TRD representation, thereby avoiding oversized features from full-sized 3D networks and the information loss of 2D networks on 2D representations. Finally, a homogeneous gated fusion network was designed to fuse the three sets of 2D features. The proposed method was evaluated on the UGRS, MIMOGR, and mmWRWD datasets. The experimental results for all datasets demonstrate that the proposed POMEN significantly and consistently outperforms the state-of-the-art models in both 2D and 3D representations.

Jigsaw puzzle solving has recently become an emerging research area. The developed techniques have been widely used in applications beyond puzzle solving. This paper focuses on solving Jigsaw Puzzles with Large Eroded Gaps (JPwLEG). We formulate the puzzle reassembly as a combinatorial optimization problem and propose a Siamese-Discriminant Deep Reinforcement Learning (SD2RL) to solve it. A Deep Q-network (DQN) is designed to visually understand the puzzles, which consists of two sets of Siamese Discriminant Networks, one set to perceive the pairwise relations between vertical neighbors and another set for horizontal neighbors. The proposed DQN considers not only the evidence from the incumbent fragment but also the support from its four neighbors. The DQN is trained using replay experience with carefully designed rewards to guide the search for a sequence of fragment swaps to reach the correct puzzle solution. Two JPwLEG datasets are constructed to evaluate the proposed method, and the experimental results show that the proposed SD2RL significantly outperforms state-of-the-art methods.

Deep neural networks facilitate video question answering (VideoQA), but the real-world applications on video streams such as CCTV and live cast place higher demands on the solver. To address the challenges of VideoQA on long videos of unknown length, we define a new set of problems called Online Open-ended Video Question Answering (O2VQA). It requires an online state-updating mechanism for the solver to decide if the collected information is sufficient to conclude an answer. We then propose a Confidence-based Event-centric Online Video Question Answering (CEO-VQA) model to solve this problem. Furthermore, a dataset called Answer Target in Background Stream (ATBS) is constructed to evaluate this newly developed online VideoQA application. Compared to the baseline VideoQA method that watches the whole video, the experimental results show that the proposed method achieves a significant performance gain.

Cross-modal learning of video and text plays a key role in Video Question Answering (VideoQA). In this paper, we propose a visual-text attention mechanism to utilize the Contrastive Language-Image Pre-training (CLIP) trained on lots of general domain language-image pairs to guide the cross-modal learning for VideoQA. Specifically, we first extract video features using a TimeSformer and text features using a BERT from the target application domain, and utilize CLIP to extract a pair of visual-text features from the general-knowledge domain through the domain-specific learning. We then propose a Cross-domain Learning to extract the attention information between visual and linguistic features across the target domain and general domain. The set of CLIP-guided visual-text features are integrated to predict the answer. The proposed method is evaluated on MSVD-QA and MSRVTTQA datasets and outperforms state-of-the-art methods.

Detecting 3D mask attacks to a face recognition system is challenging. Although genuine faces and 3D face masks show significantly different remote photoplethysmography (rPPG) signals, rPPG-based face anti-spoofing methods often suffer from performance degradation due to unstable face alignment in the video sequence and weak rPPG signals. To enhance the rPPG signal in a motion-robust way, a landmark-anchored face stitching method is proposed to align the faces robustly and precisely at the pixel-wise level by using both SIFT keypoints and facial landmarks. To better encode the rPPG signal, a weighted spatial-temporal representation is proposed, which emphasizes the face regions with rich blood vessels. In addition, characteristics of rPPG signals in different color spaces are jointly utilized. To improve the generalization capability, a lightweight EfficientNet with a Gated Recurrent Unit (GRU) is designed to extract both spatial and temporal features from the rPPG spatial-temporal representation for classification. The proposed method is compared with the state-of-the-art methods on five benchmark datasets under both intra-dataset and cross-dataset evaluations. The proposed method shows a significant and consistent improvement in performance over other state-of-the-art rPPG-based methods for face spoofing detection.

Face spoofing detection against paper attack and video-replay attack has been well studied, whereas detecting 3D face mask attack remains challenging. Remote photoplethysmography (rPPG) signal is a recently developed liveness clue for face-spoofing detection. The main challenge of existing rPPG-based methods is that the signal can be easily distorted by background noise or object motion. To address this problem, in this work, we propose an rPPG-based face-spoofing detection method using multiple regions of interests (ROIs) covering entire face, and emphasize the regions containing richer rPPG signals using larger weights. The rPPG signals of these regions form a weighted spatial-temporal map. In view of the discriminant power of EfficientNet over other deep convolutional neural networks, we propose a domain-specific EfficientNet as the classification method. Extensive experiments on two databases namely 3DMAD and HKBU-Mars V2 demonstrate the superior performance of the proposed method over state-of-the-art rPPG-based face-spoofing-detection algorithms.