Towards Explainable Action Recognition by Salient Qualitative Spatial Object Relation Chains
Authors: Hua Hua, Dongxu Li, Ruiqi Li, Peng Zhang, Jochen Renz, Anthony Cohn5710-5718
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our approach on two real-life video datasets, with respect to recognition accuracy and the quality of generated action explanations. Experiments show that our approach achieves superior performance on both aspects to previous symbolic approaches, thus facilitating trustworthy intelligent decision making. |
| Researcher Affiliation | Academia | 1 Research School of Computer Science, Australian National University, Canberra, Australia 2 School of Computing, University of Leeds, Leeds, UK {hua.hua, dongxu.li, ruiqi.li, p.zhang, jochen.renz}@anu.edu.au, a.g.cohn@leeds.ac.uk |
| Pseudocode | No | The paper describes the proposed approach and its components textually and through diagrams and equations, but it does not include a formal pseudocode or algorithm block. |
| Open Source Code | No | The paper does not include an unambiguous statement about releasing source code for the described methodology or a direct link to a code repository. |
| Open Datasets | Yes | We use two datasets in our experiments: CAD120 (Koppula, Gupta, and Saxena 2013) and CAD120++ (Zhuo et al. 2019). |
| Dataset Splits | Yes | We follow the convention as in (Tayyub et al. 2014; Zhuo et al. 2019) to divide each dataset into 4 folds based on which person performs the action (as mentioned earlier, there are 4 different actors). Namely, we apply 4-fold cross validation and results are averaged across all the 4 folds. |
| Hardware Specification | No | The paper does not provide specific details regarding the hardware specifications (e.g., GPU or CPU models) used for running the experiments. |
| Software Dependencies | No | The paper mentions the use of neural networks and state-of-the-art computer vision algorithms but does not specify any software dependencies with version numbers. |
| Experiment Setup | No | The paper describes the loss function used (cross entropy) and evaluation metrics, but it does not provide specific hyperparameter values (e.g., learning rate, batch size, number of epochs) or detailed optimizer settings for the experimental setup. |