|
| S. no | Paper title | Survey objective (existing) | Survey objective (current) |
|
| 1 | A survey of generative adversarial networks [33] | State-of-the-art GAN architectures are surveyed, and their application domains on natural language processing and computer vision are discussed. The loss functions of the GAN variants are discussed. | State-of-the-art GAN is discussed along with its performance on the MNIST dataset. Generator and discriminator losses are visually represented for the GAN variants. |
|
| 2 | Recent progress on generative adversarial networks (GANs): A survey [34] | Basic theory and different GAN models are summarized. The models derived from the GAN are classified, and evaluation metrics are discussed. | Variants of the GAN, their application, architecture, methodology, advantage, and disadvantages are analyzed and summarized. Evolution of the GAN with conditions, encoders, loss functions, and process discrete data are separately discussed. |
|
| 3 | A survey of the recent architectures of deep convolutional neural networks [35] | An overview of different layers of D-CNN, namely, the convolutional layer and pooling layer, is discussed. An outline of the pitfalls of deep learning is briefed. | Different layers of D-CNN, namely, the convolution layer, pooling layer, and the operations performed in the convolution and pooling layers, are discussed in detail. A detailed review of deep learning pitfalls, namely, overfitting, underfitting, and data insufficiency, is discussed along with their possible solutions. |
|
| 4 | Deep learning for generic object detection: A survey [36] | Recent achievements in the field of object detection have been discussed. | Recent advancements of the D-CNN in computer vision have been tabulated and discussed with their methodology and performance. Activation functions that are used for computer vision problems are tabulated. |
|
| 5 | A survey on image data augmentation for deep learning [37] | This survey presents the existing methods for data augmentation. | Advantages of data augmentation and comparing results showing the model’s performance with and without data augmentation are accomplished. |
|
| 6 | Adversarial-learning-based image-to-image transformation: A survey [38] | This survey presents an overview of adversarial learning-based methods by focusing on the image-to-image transformation scenario. | The existing survey mainly focused on image-to-image translation. This survey discusses several applications based on adversarial learning. |
|
| 7 | Survey of convolutional neural networks for image captioning [39] | This survey presents a shallow overview of image captioning performed using D-CNN. | This survey elaborately discusses various applications using the D-CNN. |
|
