The existing visual object detection methods can’t automatically adapt to illuminous changes, dynamic background due to the dynamic change of object positions in subsequent frames in video sequence. Therefore, object detection in video is a complex task. This motivates us to extract several features related to an object, including shape and texture, to detect moving objects in videos. These features employ localization and identification of the objects to improve detection accuracy in cluttered backgrounds. The texture features are obtained using a grey-level co-occurrence matrix (GLCM) and Gabor filters. Then, integrate these shape and texture features to detect the object effectively. The results are obtained on various benchmark datasets. It can handle the dynamic change of the spatial information and orientation of the object(s). Hence, the object can appear visually effectively.