Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]
Weakly-supervised Disentangling with Recurrent Transformations for 3D View Synthesis
Authors: Jimei Yang, Scott E. Reed, Ming-Hsuan Yang, Honglak Lee
NeurIPS 2015 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We carry out experiments to achieve the following objectives. First, we examine the ability of our model to synthesize high quality images of both face and complex 3D objects (chairs) in a wide range of rotational angles. Second, we evaluate the discriminative performance of disentangled identity units through cross-view object recognition. Third, we demonstrate the ability to generate and rotate novel object classes by interpolating identity units of seen objects. |
| Researcher Affiliation | Academia | Jimei Yang1 Scott Reed2 Ming-Hsuan Yang1 Honglak Lee2 1University of California, Merced EMAIL 2University of Michigan, Ann Arbor EMAIL |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access information for open-source code (e.g., repository links, explicit statements of code release, or mention of code in supplementary materials). |
| Open Datasets | Yes | Multi-PIE. The Multi-PIE [11] dataset consists of 754,204 face images from 337 people. ... Chairs. This dataset contains 1393 chair CAD models made public by Aubry et al. [1]. |
| Dataset Splits | No | The paper explicitly states training and test set splits for both datasets (e.g., 'We use the images of first 200 people for training and the remaining 137 people as the test set' for Multi-PIE, and 'We use the images of the first 500 models as the training set and the remaining 409 models as the test set' for Chairs), but it does not specify a validation set split. |
| Hardware Specification | Yes | We carry out experiments using Caffe [13] on Nvidia k40c and Titan X GPUs. |
| Software Dependencies | No | The paper mentions using 'Caffe [13]' for experiments, but it does not provide specific version numbers for Caffe or any other software dependencies needed for replication. |
| Experiment Setup | Yes | The encoder network for Multi-PIE used two convolution-relu layers with stride 2 and 2-pixel padding, followed by one fully-connected layers: 5x5x64 -> 5x5x128 -> 1024. The identity and pose units are 512 and 128, respectively. The decoder network is symmetric to the encoder... We train the network using the ADAM solver with fixed learning rate 1e-4 for 400 epochs. |