Pose and illumination variations are still the most dominating and persistent challenges haunting face recognition, leading to various highly complex 2D and 3D model-based solutions. We present a novel transform vector quantization (TVQ) method which is fast and accurate and yet much less complex than conventional methods. TVQ offers a flexible and customizable way to capture the pose variations. Use of transform such as DCT helps compressing the image data to a small feature vector and judicious use of vector quantization helps to capture the various poses into compact codebooks. A statistical confidence measure based sequence analysis allows the TVQ method to accurately recognize a person in only 3-9 frames (less than ½ a second) from a video sequence of images with wide pose variations.