Efficiently training accurate deep models for weakly supervised semantic segmentation (WSSS) with image-level labels is challenging and important. Recently, end-to-end WSSS methods have become the focus of research due to their high training efficiency. However, current methods suffer from insufficient extraction of comprehensive semantic information, resulting in low-quality pseudo-labels and sub-optimal solutions for end-to-end WSSS. To this end, we propose a simple and novel Self Correspondence Distillation (SCD) method to refine pseudo-labels without introducing external supervision. Our SCD enables the network to utilize feature correspondence derived from itself as a distillation target, which can enhance the network's feature learning process by complementing semantic information. In addition, to further improve the segmentation accuracy, we design a Variation-aware Refine Module to enhance the local consistency of pseudo-labels by computing pixel-level variation. Finally, we present an efficient end-to-end Transformer-based framework (TSCD) via SCD and Variation-aware Refine Module for the accurate WSSS task. Extensive experiments on the PASCAL VOC 2012 and MS COCO 2014 datasets demonstrate that our method significantly outperforms other state-of-the-art methods. Our code is available at https://github.com/Rongtao-Xu/RepresentationLearning/tree/main/SCD-AAAI2023.

Machine learning systems, especially the methods based on deep learning, enjoy great success in modern computer vision tasks under ideal experimental settings. Generally, these classic deep learning methods are built on the i.i.d. assumption, supposing the training and test data are drawn from the same distribution independently and identically. However, the aforementioned i.i.d. assumption is, in general, unavailable in the real-world scenarios, and as a result, leads to sharp performance decay of deep learning algorithms. Behind this, domain shift is one of the primary factors to be blamed. In order to tackle this problem, we propose using Potential Energy Ranking (PoER) to decouple the object feature and the domain feature in given images, promoting the learning of label-discriminative representations while filtering out the irrelevant correlations between the objects and the background. PoER employs the ranking loss in shallow layers to make features with identical category and domain labels close to each other and vice versa. This makes the neural networks aware of both objects and background characteristics, which is vital for generating domain-invariant features. Subsequently, with the stacked convolutional blocks, PoER further uses the contrastive loss to make features within the same categories distribute densely no matter domains, filtering out the domain information progressively for feature alignment. PoER reports superior performance on domain generalization benchmarks, improving the average top-1 accuracy by at least 1.20% compared to the existing methods. Moreover, we use PoER in the ECCV 2022 NICO Challenge, achieving top place with only a vanilla ResNet-18 and winning the jury award. The code has been made publicly available at: https://github.com/ForeverPs/PoER.

Remote sensing images are usually distributed in different departments and contain private information, so they normally cannot be available publicly. However, it is a trend to jointly use remote sensing images from different departments, because it normally enables the model to capture more information and remote sensing image analysis based on deep learning generally requires lots of training data. To address the above problem, in this paper, we apply a distributed asynchronized discriminator GAN framework (DGAN) to jointly learn remote sensing images from different client nodes. The DGAN is composed of multiple distributed discriminators and a central generator, and only the synthetic remote sensing images generated by the DGAN are used to train a semantic segmentation model. Based on DGAN, we establish an experimental platform composed of multiple different hosts, which adopts socket and multi-process technology to realize asynchronous communication between hosts, and visualize the training and testing process. During DGAN training, instead of original remote sensing images or convolutional network model information, only synthetic images, losses and labeled images are exchanged between nodes. Therefore, the DGAN well protects the privacy and security of the original remote sensing images. We verify the performance of the DGAN on three remote sensing image datasets (City-OSM, WHU and Kaggle Ship). In the experiments, we take different distributions of remote sensing images in client nodes into consideration. The experiments show that the DGAN has a great capacity for distributed remote sensing image learning without sharing the original remote sensing images or the convolutional network model. Moreover, compared with a centralized GAN trained on all remote sensing images collected from all client nodes, the DGAN can achieve almost the same performance in semantic segmentation tasks for remote sensing images.

Existing architecture semantic modeling methods in 3D complex urban scenes continue facing difficulties, such as limited training data, lack of semantic information, and inflexible model processing. Focusing on extracting and adopting accurate semantic information into a modeling process, this work presents a framework for lightweight modeling of buildings that joints point clouds semantic segmentation and 3D feature line detection constrained by geometric and photometric consistency. The main steps are: (1) Extraction of single buildings from point clouds using 2D-3D semi-supervised semantic segmentation under photometric and geometric constraints. (2) Generation of lightweight building models by using 3D plane-constrained multi-view feature line extraction and optimization. (3) Introduction of detailed semantics of building elements into independent 3D building models by using fine-grained segmentation of multi-view images to achieve high-accuracy architecture lightweight modeling with fine-grained semantic information. Experimental results demonstrate that it can perform independent lightweight modeling of each building on point cloud at various scales and scenes, with accurate geometric appearance details and realistic textures. It also enables independent processing and analysis of each building in the scenario, making them more useful in practical applications.

Machine learning systems, especially the methods based on deep learning, enjoy great success in modern computer vision tasks under experimental settings. Generally, these classic deep learning methods are built on the \emph{i.i.d.} assumption, supposing the training and test data are drawn from a similar distribution independently and identically. However, the aforementioned \emph{i.i.d.} assumption is in general unavailable in the real-world scenario, and as a result, leads to sharp performance decay of deep learning algorithms. Behind this, domain shift is one of the primary factors to be blamed. In order to tackle this problem, we propose using \textbf{Po}tential \textbf{E}nergy \textbf{R}anking (PoER) to decouple the object feature and the domain feature (\emph{i.e.,} appearance feature) in given images, promoting the learning of label-discriminative features while filtering out the irrelevant correlations between the objects and the background. PoER helps the neural networks to capture label-related features which contain the domain information first in shallow layers and then distills the label-discriminative representations out progressively, enforcing the neural networks to be aware of the characteristic of objects and background which is vital to the generation of domain-invariant features. PoER reports superior performance on domain generalization benchmarks, improving the average top-1 accuracy by at least 1.20\% compared to the existing methods. Moreover, we use PoER in the ECCV 2022 NICO Challenge\footnote{https://nicochallenge.com}, achieving top place with only a vanilla ResNet-18. The code has been made available at https://github.com/ForeverPs/PoER.