• Savitha N J
• Lata B T
• Venugopal K R

Abstract—Facial acknowledgment has gotten to be a basic innovation with differing applications, extending from security and reconnaissance to client verification and personalization. Over a long time, profound learning, especially Convolutional Neural Networks (CNNs), has revolutionized computer vision, empowering exceptional headways in facial acknowledgment assignments. This ponder dives into the synergies of consideration instruments and exchange learning inside the setting of CNNs, particularly utilizing VGGFace—a broadly recognized CNN en- gineering pre-trained on a broad dataset of facial pictures. This pre-training prepares Visual Geometry Group (VGG) Face with the capacity to capture complicated facial highlights and designs successfully. The integration of consideration components im- proves the model’s center on vital facial locales, moving forward acknowledgment accuracy. Leveraging exchange learning, the pre-trained VGGFace demonstration is fine-tuned on a particular facial acknowledgment dataset, capitalizing on the information and representations procured from the broader facial dataset. This amalgamation optimizes the precision and effectiveness of the facial acknowledgment framework, displaying the potential to assist headways in this basic innovation.

Attention Mechanism, CNN, Transfer Learning, VGGFace

"Enhancing Culprit Identification in Real-Time Video Surveillance using Deep Fine-Tuned Optimized Transfer Learning and Attention Mechanism", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 3, page no.l57-l64, March-2024, Available :http://www.jetir.org/papers/JETIR2403B07.pdf

"Enhancing Culprit Identification in Real-Time Video Surveillance using Deep Fine-Tuned Optimized Transfer Learning and Attention Mechanism", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 3, page no. ppl57-l64, March-2024, Available at : http://www.jetir.org/papers/JETIR2403B07.pdf