Head pose estimation is a fundamental task in facial analysis and behavior understanding. To address the limitations of 6DRepNet in single-RGB scenarios, particularly in high-level spatial discriminative region modeling, global feature-to-6D rotation representation mapping, and the optimization of large-pose challenging samples, this paper proposes a geometry-aware enhanced framework for head pose estimation. While preserving the 6D continuous rotation representation and its SO(3)-based geometric supervision mechanism, the proposed method improves the baseline model through the joint design of a Spatial Recalibration Module, a Residual Pose Mapping Head, and a Pose-Aware Weighted Geodesic Loss. Experiments are conducted using 300W-LP for training and AFLW2000 and BIWI for evaluation. The results show that the proposed method consistently outperforms the baseline 6DRepNet on both datasets, reducing the overall MAE from 3.97 to 3.72 on AFLW2000 and from 3.54 to 3.26 on BIWI. Ablation studies further verify the effectiveness and complementarity of the proposed components. These results demonstrate that the proposed method can effectively improve the accuracy and robustness of single-RGB head pose estimation without introducing additional modalities.

Multimodal sentiment analysis must combine textual, visual, and acoustic evidence whose reliability varies across utterances. We propose PGMCR, which uses a predicted three-way polarity distribution to compose sample-level modality-reliability weights and applies a training-only consistency constraint to text-preserving visual and acoustic perturbations. Experiments on CMU-MOSI and CMU-MOSEI show that polarity-guided fusion is the principal architectural contribution, while the text-guided refinement and consistency branch have metric-dependent effects. The learned weights vary across polarity groups, but proxy-corruption diagnostics also reveal strong dependence on textual evidence and non-negligible polarity-estimation errors. The results support polarity as a useful conditioning signal for multimodal fusion while delimiting the scope of the nonverbal robustness claim.

Despite significant advances, multimodal sentiment analysis still faces critical challenges in modeling complex cross-modal interactions and extracting discriminative sentiment features. To address these limitations, this paper proposes a hierarchical multimodal sentiment analysis framework. Specifically, a cross-modal feature enhancement module is first introduced to capture deep correlations among textual, visual, and acoustic modalities via cross-attention mechanisms, thereby obtaining context-aware fused representations. Subsequently, an attention-gated feature disentanglement approach is employed to effectively separate sentiment-relevant information from content-specific features within the fused representations; an independence loss is further imposed to enforce orthogonality between these two feature subsets, thereby mitigating noise induced by repetitive visual frames and textual stop words. Finally, all disentangled features are integrated to facilitate high-level sentiment reasoning through a multimodal logical inference module, where supervised contrastive loss is incorporated to enhance the discriminability of sentiment expressions. Extensive experiments conducted on two public benchmarks, CMU-MOSI and CMU-MOSEI, demonstrate that the proposed framework achieves improvements of 2–6% across multiple evaluation metrics compared with state-of-the-art methods.