3D Concept Grounding on Neural Fields

Authors: Yining Hong, Yilun Du, Chunru Lin, Josh Tenenbaum, Chuang Gan

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on our collected PARTNET-REASONING dataset show that our proposed framework outperforms unsupervised / language-mediated segmentation models on semantic and instance segmentation tasks, as well as outperforms existing models on the challenging 3D aware visual reasoning tasks.
Researcher Affiliation Collaboration Yining Hong University of California, Los Angeles Yilun Du Massachusetts Institute of Technology Chunru Lin Shanghai Jiao Tong University Joshua B. Tenenbaum MIT BCS, CBMM, CSAIL Chuang Gan UMass Amherst and MIT-IBM Watson AI Lab
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper states in its self-assessment: '3. If you ran experiments... (a) Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [No]' and '4. If you are using existing assets (e.g., code, data, models) or curating/releasing new assets... (c) Did you include any new assets either in the supplemental material or as a URL)? [No]'
Open Datasets Yes To this end, we collect a new dataset, Part Net-Reasoning, which focuses on visual question answering on the Part Net [31] dataset. Specifically, we render approximately 3K RGB-D images from shapes of 4 categories: Chair, Table, Bag and Cart, with 8 question-answer pairs on average for each shape.
Dataset Splits No The paper mentions collecting a dataset and a curriculum learning strategy for training but does not provide specific details on training, validation, or test splits (e.g., percentages or sample counts).
Hardware Specification No The paper does not provide specific details about the hardware (e.g., GPU model, CPU, memory) used to run the experiments.
Software Dependencies No The paper does not list specific software dependencies with their version numbers required for replication.
Experiment Setup No The paper mentions joint optimization and training for a certain number of epochs (N1) and using parameters alpha and beta in the loss function, but it does not specify the numerical values for these hyperparameters (e.g., learning rate, batch size, N1, alpha, beta) or other detailed training configurations.