This paper explores and compares various techniques for automatically recognizing facial actions in image sequences. The Facial Action Coding System (FACS) is a widely used method for quantifying facial movements, but it is time-consuming and requires highly trained human experts. The paper investigates methods such as optical flow analysis, holistic spatial analysis (using principal component analysis, independent component analysis, local feature analysis, and linear discriminant analysis), and local spatial analysis (using Gabor wavelets and local principal components). The performance of these systems is compared to that of naive and expert human subjects. The best results were obtained using Gabor wavelet representation and independent component representation, both achieving 96% accuracy in classifying 12 facial actions of the upper and lower face. The findings highlight the importance of using local filters, high spatial frequencies, and statistical independence for effective facial action classification.This paper explores and compares various techniques for automatically recognizing facial actions in image sequences. The Facial Action Coding System (FACS) is a widely used method for quantifying facial movements, but it is time-consuming and requires highly trained human experts. The paper investigates methods such as optical flow analysis, holistic spatial analysis (using principal component analysis, independent component analysis, local feature analysis, and linear discriminant analysis), and local spatial analysis (using Gabor wavelets and local principal components). The performance of these systems is compared to that of naive and expert human subjects. The best results were obtained using Gabor wavelet representation and independent component representation, both achieving 96% accuracy in classifying 12 facial actions of the upper and lower face. The findings highlight the importance of using local filters, high spatial frequencies, and statistical independence for effective facial action classification.