This study evaluates frame selection during inference as an independent factor in video deepfake detection while keeping the downstream detectors fixed. We compare twelve frame selection strategies, ranging from simple temporal and quality baselines to landmark aware policies, using four validated pretrained detectors: Self-Blended Images (SBIs), Frequency-Enhanced Self-Blended Images (FSBIs), Generative Convolutional Vision Transformer (GenConViT), and GenD. The primary experiment is a complete factorial benchmark with 300 videos and five frame budgets (2, 4, 8, 16, and 32 selected frames), which provides the reference results at 32 frames. To address sample size limitations, an additional validation experiment uses a deduplicated split of 1180 Celeb-DF++ and FaceForensics++ videos, with complete results for 2, 4, and 8 selected frames and a reported subset for 16 selected frames. In the complete 300-video benchmark, 32 frames achieved the strongest average AUC, while 8 and 16 frames recovered most of the attainable performance with lower runtime. The best single validated configuration was GenD with Shot-aware sampling at 32 frames, yielding an AUC of 0.9607 and a balanced accuracy of 0.9133. The study therefore does not claim that smaller budgets universally outperform 32 frames; instead, it quantifies the tradeoff between accuracy and runtime and shows that frame selection remains a meaningful design variable under constrained inference budgets.

Pain episode detection and forecasting from physiological signals collected by wearable devices is becoming increasingly important. Self-reported pain ratings in daily life are recorded irregularly, leaving episode onset, duration, and dynamics incompletely documented. The reviewed literature shows that physiological changes are driven not only by pain but also by movement, emotional arousal, sleep, and environmental factors. Therefore, real-world reliability and false-episode reduction remain key obstacles. This focused review synthesizes clinical and naturalistic use cases, data acquisition strategies, and modeling approaches. These approaches range from feature engineering and classical machine learning to multimodal fusion with attention mechanisms and transformer architectures. Beyond summarizing established findings from the literature, the paper adds two literature-derived synthesis recommendations. The 2 + 1 rule treats at least two physiological signal families plus one context channel as a practical design heuristic for robust wearable pain detection. The KKS principle uses quality indicators and context information to control fusion weights dynamically. Across the reviewed studies, single-signal solutions are frequently vulnerable to signal-quality fluctuations. By contrast, hybrid strategies with explicit quality and context handling show stronger translational potential. However, these recommendations should be interpreted as conceptual guidance requiring further empirical validation, especially across participants, devices, protocols, and uncontrolled daily-life conditions.

Deepfake detection from images and videos has evolved from artifact-specific convolutional baselines toward more generalizable, cross-dataset, and foundation-model-based approaches. The current work focuses on the efficiency and informativeness of frame selection itself, while keeping the downstream detectors fixed. The study compares twelve frame-selection heuristics ranging from simple baselines to landmark-aware strategies. Four pre-trained detectors were included in the present quantitative comparison: Self-Blended Images (SBI), Frequency-Enhanced Self-Lendered Images (FSBI), Generative Convolutional Vision Transformer (GenConViT), and GenD. The results show that GenD achieved the strongest average detector-level performance, with a mean frame-mean AUC of 0.9464. The best single validated configuration is GenD, yielding an AUC value of 0.9607 and a balanced accuracy of 0.9133. FSBI and SBI reached mean AUC values of 0.8953 and 0.8935, respectively, while GenD was the best general candidate. For SBI, the best validation configuration is Landmark cluster with 32 selected frames. GenD achieves the best AUC at the level of selection strategy. The present work demonstrates that inference-time frame selection is an important component of video-only deepfakes under constrained inference budgets.