Scanpath Prediction for Visual Attention using IOR-ROI LSTM

Authors: Zhenzhong Chen, Wanjie Sun

IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results indicate that the proposed architecture can achieve superior performance in predicting scanpaths. Results on the OSIE dataset [Xu et al., 2014] and MIT dataset [Judd et al., 2009] have verified the effectiveness of our model. We trained and evaluated our proposed model on the widely used public OSIE and MIT eye-tracking datasets.
Researcher Affiliation Academia Zhenzhong Chen, Wanjie Sun School of Remote Sensing and Information Engineering, Wuhan University, Wuhan, P.R. China
Pseudocode No The paper describes the model architecture and equations but does not include structured pseudocode or an algorithm block.
Open Source Code No The paper does not provide any statement about releasing source code or a link to a code repository.
Open Datasets Yes We trained and evaluated our proposed model on the widely used public OSIE and MIT eye-tracking datasets consisting of natural images along with eye-tracking data from different participants. In the OSIE dataset, there are 700 images with 800 600 pixels. Eye-tracking data was acquired from 15 participants for each image. The MIT dataset is composed of 1003 images with resolution ranging from 405 to 1024. Eye-tracking data was also recorded from 15 subjects for each image.
Dataset Splits Yes We randomly splitted the OSIE and MIT datasets into 80% training data and 20% test data.
Hardware Specification No The paper mentions implementing parts "on the modern GPU" and using a "pre-trained VGG-19 network" but does not specify any exact GPU or CPU models, memory sizes, or other detailed hardware specifications used for running the experiments.
Software Dependencies No The paper mentions various components like Conv LSTM, VGG-19, Adam optimizer, and ReLU activation functions but does not specify any software names with version numbers (e.g., Python, TensorFlow, PyTorch versions, specific libraries).
Experiment Setup Yes The model was trained using the Adam optimizer along with step learning rate decay strategy which decays learning rate at a constant rate of 0.9. Initial learning rate for the IORROI LSTM and saliency guidance part is 1 10 4 while fine-tuning the dilated VGG-19 with an initial learning rate of 1 10 5.