Two-view correspondence pruning aims to identify reliable correspondences for camera pose estimation, serving as a fundamental step in many 3D vision tasks. Existing methods rely on geometric consistency to seek true correspondences (inliers) from numerous false correspondences (outliers). In this learning paradigm, outliers severely affect the representation learning of inliers, resulting in models that are neither robust nor generalizable. To address this issue, we propose a geometry-consistent pre-training paradigm that sculpts scalable and generalizable representations free from outlier interference. The paradigm features two appealing properties. 1) Implementation of geometry-consistent pre-training. We introduce masked inlier reconstruction as a pretext task and develop a simple yet effective pre-training framework based on a masked autoencoder. Specifically, due to the irregular and unordered nature of correspondences, which lack explicit positional information, we adopt a dual-branch structure that separately reconstructs the keypoints of two images. This enables indirect reconstruction of 4D correspondences, where keypoints from the paired image provide positional prompts. 2) Unified correspondence encoder. We propose a simple dual-stream encoder with built-in consensus interaction, providing a unified, extensible architecture that enhances representation learning. Extensive experiments demonstrate that our method, GeneralPruner, consistently outperforms state-of-the-art approaches in terms of robustness and generalization across various downstream tasks. Specifically, our method achieves 10.76%, 11.84%, and 8.65% performance gains in camera pose estimation, visual localization, and 3D registration, respectively. To the best of our knowledge, we are the first work to introduce a pre-training framework tailored for correspondence pruning, offering a more universal and scalable solution.

Among adversarial attacks against sequential recommenders, model extraction attacks represent a promising method, aiming to a construct surrogate model for downstream attacks, such as data poisoning attack and profile pollution attack. Existing research on model extraction neglects the role of few-shot data in improving the performance of the surrogate model. When an adversary has access to few-shot raw data, how to maximize the utilization of this limited data to construct a surrogate recommender that more closely approximates the victim recommender, remains an open issue. In this paper, we propose FewMEA, a framework that utilizes only a minimal amount of raw data (less than 10%) yet significantly enhances the similarity of model outputs. In FewMEA, we introduce a novel method for generating synthetic data, ensuring the generated data more closely aligns with the distribution of the raw data. We also design a novel loss function that effectively reduces the discrepancy between the surrogate model and the victim model by focusing on the differences in their recommendation lists. Experiments on three datasets demonstrate that the proposed FewMEA significantly enhances output similarity, achieving an average improvement of 18.62% compared to state-of-the-art model extraction frameworks against sequential recommenders.

Pyramid Temporal Hierarchy Network (PTH-Net) is a new paradigm for dynamic facial expression recognition, applied directly to raw videos, without face detection and alignment. Unlike the traditional paradigm, which focus only on facial areas and often overlooks valuable information like body movements, PTH-Net preserves more critical information. It does this by distinguishing between backgrounds and human bodies at the feature level, offering greater flexibility as an end-to-end network. Specifically, PTH-Net utilizes a pre-trained backbone to extract multiple general features of video understanding at various temporal frequencies, forming a temporal feature pyramid. It then further expands this temporal hierarchy through differentiated parameter sharing and downsampling, ultimately refining emotional information under the supervision of expression temporal-frequency invariance. Additionally, PTH-Net features an efficient Scalable Semantic Distinction layer that enhances feature discrimination, helping to better identify target expressions versus non-target ones in the video. Finally, extensive experiments demonstrate that PTH-Net performs excellently in eight challenging benchmarks, with lower computational costs compared to previous methods. The source code is available at https://github.com/lm495455/PTH-Net.

Dynamic expression recognition in the wild is a challenging task due to various obstacles, including low light condition, non-positive face, and face occlusion. Purely vision-based approaches may not suffice to accurately capture the complexity of human emotions. To address this issue, we propose a Transformer-based Multimodal Emotional Perception (T-MEP) framework capable of effectively extracting multimodal information and achieving significant augmentation. Specifically, we design three transformer-based encoders to extract modality-specific features from audio, image, and text sequences, respectively. Each encoder is carefully designed to maximize its adaptation to the corresponding modality. In addition, we design a transformer-based multimodal information fusion module to model cross-modal representation among these modalities. The unique combination of self-attention and cross-attention in this module enhances the robustness of output-integrated features in encoding emotion. By mapping the information from audio and textual features to the latent space of visual features, this module aligns the semantics of the three modalities for cross-modal information augmentation. Finally, we evaluate our method on three popular datasets (MAFW, DFEW, and AFEW) through extensive experiments, which demonstrate its state-of-the-art performance. This research offers a promising direction for future studies to improve emotion recognition accuracy by exploiting the power of multimodal features.