• MEGHA AGARWAL

In this paper, we propose a novel facial expression recognition framework using bimodal information. Our contributions are two folds. First, our frame-work explicitly models the cross-modality data correlation while allowing them to be treated as asynchronous streams. Second, we generate a single image representing the key emotion in the video (image sequence) containing hundreds of frames. This helps to circumvent the complex and noisy dynamics in the video frames and at the same time enables to utilize image based facial expression approaches. We also show that by incorporating cross-modal information a significant reduction in computation cost can be achieved. On the other hand by avoiding spurious frames for further processing and thereby reducing unwanted influence, classification accuracy can be improved.

"AUTOMATIC FACE EMOTION RECOGNITION BY AUDIO-VISUAL CROSS MODEL DATA ASSOCIATION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.669-671, June 2019, Available :http://www.jetir.org/papers/JETIR1907399.pdf

"AUTOMATIC FACE EMOTION RECOGNITION BY AUDIO-VISUAL CROSS MODEL DATA ASSOCIATION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp669-671, June 2019, Available at : http://www.jetir.org/papers/JETIR1907399.pdf