Continuous Surface Embeddings

Authors: Natalia Neverova, David Novotny, Marc Szafraniec, Vasil Khalidov, Patrick Labatut, Andrea Vedaldi

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate that the proposed approach performs on par or better than the state-of-the-art methods for dense pose estimation for humans, while being conceptually simpler. We also collect a new in-the-wild dataset of dense correspondences for animal classes and demonstrate that our framework scales naturally to the new deformable object categories. 5 Experiments. The experimental results are reported following the updated protocol based on GPSm scores [54].
Researcher Affiliation Industry Natalia Neverova David Novotny Vasil Khalidov Marc Szafraniec Patrick Labatut Andrea Vedaldi {nneverova, dnovotny, vkhalidov, mszafraniec, plabatut, vedaldi} fb.com. Facebook AI Research
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The code, trained models and the dataset will be made publicly available to ensure reproducibility.
Open Datasets Yes We rely on the Dense Pose-COCO dataset [17] for evaluation of the proposed method on the human category and comparison with the Dense Pose (IUV) training. Additionally, we collect correspondence annotations on a set of 9 animal categories of the LVIS dataset [21]. Based on images from the COCO dataset [31], LVIS features significantly more accurate object masks.
Dataset Splits Yes The training is performed on 8 GPUs for 130k iterations on Dense Pose-COCO (standard s1x schedule [54]) and 5k iterations on Dense Pose-Chimps and Dense Pose-LVIS. Table 2: Performance on Dense Pose-COCO, with IUV (top) and CSE (bottom) training (GPSm scores, minival). Table 3: Hyperparameter search and performance in low data regimes (AP, Dense Pose COCO, minival).
Hardware Specification No The training is performed on 8 GPUs. This is a general statement and does not specify the model or type of GPUs, or any other hardware components.
Software Dependencies No Our networks are implemented in PyTorch within the Detectron2 [54] framework. The paper mentions software names but does not provide specific version numbers for reproducibility.
Experiment Setup Yes The training is performed on 8 GPUs for 130k iterations on Dense Pose-COCO (standard s1x schedule [54]) and 5k iterations on Dense Pose-Chimps and Dense Pose-LVIS. All CSE models are trained with loss Lσ (eq. 4), LBO size M = 256, embedding size D = 16.