Action-Conditioned Generation of Bimanual Object Manipulation Sequences
Authors: Haziq Razali, Yiannis Demiris
AAAI 2023 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our approach on the KIT Motion Capture and KIT RGBD Bimanual Manipulation datasets and show improvements over a simplified approach that treats the entire body as a single entity, and existing whole-body-only methods. |
| Researcher Affiliation | Academia | Haziq Razali, Yiannis Demiris* Personal Robotics Lab, Dept. of Electrical and Electronic Engineering, Imperial College London EMAIL |
| Pseudocode | No | The paper describes the method using text and equations but does not include structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Code at www.imperial.ac.uk/personal-robotics/software |
| Open Datasets | Yes | Datasets The KIT Motion Capture Dataset (Krebs et al. 2021) contains motion capture data of a right-handed person performing bimanual actions such as Cut, Pour, Stir, etc.The KIT RGBD Dataset (Dreher, W achter, and Asfour 2019) differs in that it contains RGBD recordings instead of motion capture. |
| Dataset Splits | No | The paper mentions "train:test ratio of 70:30" but does not explicitly state a separate validation split or its proportions. |
| Hardware Specification | Yes | The experiments were implemented using Py Torch 1.12.0 installed on an Ubuntu 20.04 machine with an NVIDIA RTX-2080. |
| Software Dependencies | Yes | The experiments were implemented using Py Torch 1.12.0 installed on an Ubuntu 20.04 machine with an NVIDIA RTX-2080. |
| Experiment Setup | Yes | We train all models including ours for 1000 epochs using the ADAM optimizer (Kingma and Ba 2014) with an initial learning rate of 1e-3 and batch size of 32. |