MIMIC: Masked Image Modeling with Image Correspondences
- URL: http://arxiv.org/abs/2306.15128v4
- Date: Thu, 16 May 2024 03:03:37 GMT
- Title: MIMIC: Masked Image Modeling with Image Correspondences
- Authors: Kalyani Marathe, Mahtab Bigverdi, Nishat Khan, Tuhin Kundu, Patrick Howe, Sharan Ranjit S, Anand Bhattad, Aniruddha Kembhavi, Linda G. Shapiro, Ranjay Krishna,
- Abstract summary: Current methods for building effective pretraining datasets rely on annotated 3D meshes, point clouds, and camera parameters from simulated environments.
We propose a pretraining dataset-curation approach that does not require any additional annotations.
Our method allows us to generate multi-view datasets from both real-world videos and simulated environments at scale.
- Score: 29.8154890262928
- License: http://creativecommons.org/licenses/by-nc-sa/4.0/
- Abstract: Dense pixel-specific representation learning at scale has been bottlenecked due to the unavailability of large-scale multi-view datasets. Current methods for building effective pretraining datasets heavily rely on annotated 3D meshes, point clouds, and camera parameters from simulated environments, preventing them from building datasets from real-world data sources where such metadata is lacking. We propose a pretraining dataset-curation approach that does not require any additional annotations. Our method allows us to generate multi-view datasets from both real-world videos and simulated environments at scale. Specifically, we experiment with two scales: MIMIC-1M with 1.3M and MIMIC-3M with 3.1M multi-view image pairs. We train multiple models with different masked image modeling objectives to showcase the following findings: Representations trained on our automatically generated MIMIC-3M outperform those learned from expensive crowdsourced datasets (ImageNet-1K) and those learned from synthetic environments (MULTIVIEW-HABITAT) on two dense geometric tasks: depth estimation on NYUv2 (1.7%), and surface normals estimation on Taskonomy (2.05%). For dense tasks which also require object understanding, we outperform MULTIVIEW-HABITAT, on semantic segmentation on ADE20K (3.89%), pose estimation on MSCOCO (9.4%), and reduce the gap with models pre-trained on the object-centric expensive ImageNet-1K. We outperform even when the representations are frozen, and when downstream training data is limited to few-shot. Larger dataset (MIMIC-3M) significantly improves performance, which is promising since our curation method can arbitrarily scale to produce even larger datasets. MIMIC code, dataset, and pretrained models are open-sourced at https://github.com/RAIVNLab/MIMIC.
Related papers
- MMScan: A Multi-Modal 3D Scene Dataset with Hierarchical Grounded Language Annotations [55.022519020409405]
This paper builds the first largest ever multi-modal 3D scene dataset and benchmark with hierarchical grounded language annotations, MMScan.
The resulting multi-modal 3D dataset encompasses 1.4M meta-annotated captions on 109k objects and 7.7k regions as well as over 3.04M diverse samples for 3D visual grounding and question-answering benchmarks.
arXiv Detail & Related papers (2024-06-13T17:59:30Z) - MMEarth: Exploring Multi-Modal Pretext Tasks For Geospatial Representation Learning [9.540487697801531]
MMEarth is a diverse multi-modal pretraining dataset at global scale.
We propose a Multi-Pretext Masked Autoencoder (MP-MAE) approach to learn general-purpose representations for optical satellite images.
arXiv Detail & Related papers (2024-05-04T23:16:48Z) - Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Image [85.91935485902708]
We show that the key to a zero-shot single-view metric depth model lies in the combination of large-scale data training and resolving the metric ambiguity from various camera models.
We propose a canonical camera space transformation module, which explicitly addresses the ambiguity problems and can be effortlessly plugged into existing monocular models.
Our method enables the accurate recovery of metric 3D structures on randomly collected internet images.
arXiv Detail & Related papers (2023-07-20T16:14:23Z) - Delving Deeper into Data Scaling in Masked Image Modeling [145.36501330782357]
We conduct an empirical study on the scaling capability of masked image modeling (MIM) methods for visual recognition.
Specifically, we utilize the web-collected Coyo-700M dataset.
Our goal is to investigate how the performance changes on downstream tasks when scaling with different sizes of data and models.
arXiv Detail & Related papers (2023-05-24T15:33:46Z) - DiM: Distilling Dataset into Generative Model [42.32433831074992]
We propose a novel distillation scheme to textbfDistill information of large train sets textbfinto generative textbfModels, named DiM.
During the distillation phase, we minimize the differences in logits predicted by a models pool between real and generated images.
At the deployment stage, the generative model synthesizes various training samples from random noises on the fly.
arXiv Detail & Related papers (2023-03-08T16:48:24Z) - Towards Multimodal Multitask Scene Understanding Models for Indoor
Mobile Agents [49.904531485843464]
In this paper, we discuss the main challenge: insufficient, or even no, labeled data for real-world indoor environments.
We describe MMISM (Multi-modality input Multi-task output Indoor Scene understanding Model) to tackle the above challenges.
MMISM considers RGB images as well as sparse Lidar points as inputs and 3D object detection, depth completion, human pose estimation, and semantic segmentation as output tasks.
We show that MMISM performs on par or even better than single-task models.
arXiv Detail & Related papers (2022-09-27T04:49:19Z) - MSeg: A Composite Dataset for Multi-domain Semantic Segmentation [100.17755160696939]
We present MSeg, a composite dataset that unifies semantic segmentation datasets from different domains.
We reconcile the generalization and bring the pixel-level annotations into alignment by relabeling more than 220,000 object masks in more than 80,000 images.
A model trained on MSeg ranks first on the WildDash-v1 leaderboard for robust semantic segmentation, with no exposure to WildDash data during training.
arXiv Detail & Related papers (2021-12-27T16:16:35Z) - Multimodal Semi-Supervised Learning for 3D Objects [19.409295848915388]
This paper explores how the coherence of different modelities of 3D data can be used to improve data efficiency for both 3D classification and retrieval tasks.
We propose a novel multimodal semi-supervised learning framework by introducing instance-level consistency constraint and a novel multimodal contrastive prototype (M2CP) loss.
Our proposed framework significantly outperforms all the state-of-the-art counterparts for both classification and retrieval tasks by a large margin on the modelNet10 and ModelNet40 datasets.
arXiv Detail & Related papers (2021-10-22T05:33:16Z) - M3DeTR: Multi-representation, Multi-scale, Mutual-relation 3D Object
Detection with Transformers [78.48081972698888]
We present M3DeTR, which combines different point cloud representations with different feature scales based on multi-scale feature pyramids.
M3DeTR is the first approach that unifies multiple point cloud representations, feature scales, as well as models mutual relationships between point clouds simultaneously using transformers.
arXiv Detail & Related papers (2021-04-24T06:48:23Z) - Generating synthetic photogrammetric data for training deep learning
based 3D point cloud segmentation models [0.0]
At I/ITSEC 2019, the authors presented a fully-automated workflow to segment 3D photogrammetric point-clouds/meshes and extract object information.
The ultimate goal is to create realistic virtual environments and provide the necessary information for simulation.
arXiv Detail & Related papers (2020-08-21T18:50:42Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.