Mujtaba Asad

TL;DR: Wang et al. as discussed by the authors proposed a 3D Convolutional Autoencoder (3D-CAE) architecture to extract spatio-temporal features from normal event training videos, which can learn appearance and motion features effectively in an unsupervised manner.

TL;DR: An image detail enhancement algorithm based on in-place residual homogeneity (IP), which gets the detail layer directly and amplifies it and has many good properties, such as being fast, edge-aware, robust, and parameter-free.

TL;DR: Experimental results on three different standard benchmark datasets show that the proposed algorithm provides better ability to recognize violent actions in different scenarios and results in improved performance compared to the state-of-the-art methods.

TL;DR: This paper proposes an end-to-end zero-shot sketch-based image retrieval approach using an indicative Vision Transformer, addressing limitations of existing methods by incorporating a feature picker, parallel feature adapter, and logit-level auxiliary signal for improved performance on various datasets.