Cross-Attentional Audio-Visual Fusion for Weakly-Supervised Action Localization

Authors: Jun-Tae Lee, Mihir Jain, Hyoungwoo Park, Sungrack Yun

ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on two publicly available video-datasets (AVE and Activity Net1.2) show that the proposed method effectively fuses audio and visual modalities, and achieves the state-of-the-art results for weakly-supervised action localization.
Researcher Affiliation Industry Juntae Lee, Mihir Jain, Hyoungwoo Park & Sungrack Yun Qualcomm AI Research . {juntlee,mijain,hwoopark,sungrack}@qti.qualcomm.com
Pseudocode No The paper does not contain any clearly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not include a statement about releasing its source code or provide a link to a code repository for the methodology described.
Open Datasets Yes Extensive experiments are conducted on two video datasets for localizing audio-visual events (AVE1) and actions (Activity Net1.22). 1https://github.com/Yapeng Tian/AVE-ECCV18 2http://activity-net.org/download.html
Dataset Splits Yes Activity Net1.2 is a temporal action localization dataset with 4,819 train and 2,383 validation videos, which in the literature is used for evaluation.
Hardware Specification No The paper does not explicitly describe the specific hardware (e.g., GPU models, CPU types, memory) used to run its experiments.
Software Dependencies No The paper mentions using specific networks like I3D, ResNet152, and VGG-like network for feature extraction, but it does not specify versions for software libraries or dependencies (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup Yes We set dx to 1,024, and the Leaky Relu and hyperbolic tangent functions are respectively used for the activation of modality-specific layers and cross-attention modules. In training, the parameters are initialized with Xavier method (Glorot & Bengio, 2010) and updated by Adam optimizer (Kingma & Ba, 2015) with the learning rate of 10 4 and the batch size of 30. Also, the dropout with a ratio of 0.7 is applied for the final attended audio-visual features. In the loss, the hyper parameters are set as B = 4, α = 0.8, β = 0.8 and γ = 1.