• Corpus ID: 247084335

FreeSOLO: Learning to Segment Objects without Annotations

@article{Wang2022FreeSOLOLT,
  title={FreeSOLO: Learning to Segment Objects without Annotations},
  author={Xinlong Wang and Zhiding Yu and Shalini De Mello and Jan Kautz and Anima Anandkumar and Chunhua Shen and Jos{\'e} Manuel {\'A}lvarez},
  journal={ArXiv},
  year={2022},
  volume={abs/2202.12181}
}
Instance segmentation is a fundamental vision task that aims to recognize and segment each object in an image. However, it requires costly annotations such as bounding boxes and segmentation masks for learning. In this work, we propose a fully unsupervised learning method that learns class-agnostic instance segmentation without any annotations. We present FreeSOLO, a self-supervised instance segmentation framework built on top of the simple instance segmentation method SOLO. Our method also… 

Figures and Tables from this paper

Anomaly Detection in Autonomous Driving: A Survey
TLDR
This survey provides an extensive overview of anomaly detection techniques based on camera, lidar, radar, multimodal and abstract object level data and outlines the state-of-the-art and point out current research gaps.
MinVIS: A Minimal Video Instance Segmentation Framework without Video-based Training
TLDR
MinVIS has the practical advantages of reducing both the labeling costs and the memory requirements, while not sacrificing the VIS performance, and is comparable to fully-supervised state-of-the-art approaches on YouTube-VIS 2019/2021.

References

SHOWING 1-10 OF 77 REFERENCES
SOLO: A Simple Framework for Instance Segmentation
TLDR
This paper introduces the notion of “instance categories”, which assigns categories to each pixel within an instance according to the instance’s location, and proposes segmenting objects by locations (SOLO), a simple, direct, and fast framework for instance segmentation with strong momentum.
SOLO: Segmenting Objects by Locations
TLDR
A new, embarrassingly simple approach to instance segmentation in images by introducing the notion of "instance categories", which assigns categories to each pixel within an instance according to the instance's location and size thus nicely converting instance mask segmentation into a classification-solvable problem.
Pointly-Supervised Instance Segmentation
TLDR
The existing instance segmentation models developed for full mask supervision can be seamlessly trained with point-based supervision collected via the authors' scheme, and the new module, called Implicit PointRend, is more straightforward and uses a single point-level mask loss.
SegSort: Segmentation by Discriminative Sorting of Segments
TLDR
The SegSort is presented, as a first attempt using deep learning for unsupervised semantic segmentation, achieving 76% performance of its supervised counterpart and producing an interpretable result.
Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals
TLDR
A two-step framework that adopts a predetermined mid-level prior in a contrastive optimization objective to learn pixel embeddings and argues about the importance of having a prior that contains information about objects, or their parts, and discusses several possibilities to obtain such a prior in an unsupervised manner.
Localizing Objects with Self-Supervised Transformers and no Labels
TLDR
This work proposes a simple approach to object discovery, that leverages the activation features of a vision transformer pre-trained in a self-supervised manner, that outperform state-of-the-art object discovery methods by up to 8 CorLoc points on PASCAL VOC 2012.
Show, Match and Segment: Joint Weakly Supervised Learning of Semantic Matching and Object Co-Segmentation
TLDR
The model is end-to-end trainable and does not require supervision from manually annotated correspondences and object masks, and performs favorably against the state-of-the-art methods on both semantic matching and object co-segmentation tasks.
Leveraging Instance-, Image- and Dataset-Level Information for Weakly Supervised Instance Segmentation
TLDR
This article proposes a multiple instance learning (MIL) framework, which can be trained in an end-to-end manner using training images with image-level labels and achieves state-of-the-art performance for both weakly supervised instance segmentation and semantic segmentation.
SOLOv2: Dynamic and Fast Instance Segmentation
TLDR
State-of-the-art results in object detection (from the authors' mask byproduct) and panoptic segmentation show the potential to serve as a new strong baseline for many instance-level recognition tasks besides instance segmentation.
Semantic Instance Segmentation with a Discriminative Loss Function
TLDR
This work proposes an approach of combining an off-the-shelf network with a principled loss function inspired by a metric learning objective that encourages a convolutional network to produce a representation of the image that can easily be clustered into instances with a simple post-processing step.
...
...