Abstract: Unsupervised domain adaptation(UDA) has been applied to image semantic segmentation to solve the problem of domain offset. However, in some difficult categories with poor recognition accuracy, the segmentation effects are still not ideal. To this end, in this paper, Intra-subdomain adaptation adversarial learning segmentation method based on Dynamic Pseudo Labels(IDPL) is proposed. The whole process consists of 3 steps: Firstly, the instance-level pseudo label dynamic generation module is proposed, which fuses the class matching information in global classes and local instances, thus adaptively generating the optimal threshold for each class, obtaining high-quality pseudo labels. Secondly, the subdomain classifier module based on instance confidence is constructed, which can dynamically divide the target domain into easy and difficult subdomains according to the relative proportion of easy and difficult instances. Finally, the subdomain adversarial learning module based on self-attention is proposed.

Object detection algorithms allow to enable many interesting applications which can be implemented in different devices, such as smartphones and wearable devices. In the context of a cultural site, implementing these algorithms in a wearable device, such as a pair of smart glasses, allow to enable the use of augmented reality (AR) to show extra information about the artworks and enrich the visitors' experience during their tour. However, object detection algorithms require to be trained on many well annotated examples to achieve reasonable results. This brings a major limitation since the annotation process requires human supervision which makes it expensive in terms of time and costs. A possible solution to reduce these costs consist in exploiting tools to automatically generate synthetic labeled images from a 3D model of the site. However, models trained with synthetic data do not generalize on real images acquired in the target scenario in which they are supposed to be used. Furthermore, object detectors should be able to work with different wearable devices or different mobile devices, which makes generalization even harder. In this paper, we present a new dataset collected in a cultural site to study the problem of domain adaptation for object detection in the presence of multiple unlabeled target domains corresponding to different cameras and a labeled source domain obtained considering synthetic images for training purposes. We present a new domain adaptation method which outperforms current state-of-the-art approaches combining the benefits of aligning the domains at the feature and pixel level with a self-training process. We release the dataset at the following link https://iplab.dmi.unict.it/OBJ-MDA/ and the code of the proposed architecture at https://github.com/fpv-iplab/STMDA-RetinaNet.

WASHINGTON – U.S. Defense Secretary Mark Esper said on Tuesday the United States wants to de-escalate tensions with Iran, but the country is ready to finish any war that could be started. "We are not looking to start a war with Iran, but we are prepared to finish one," he said. "What we'd like to see is the situation de-escalated." The U.S. drone strike on Friday that killed Iranian military commander Qassem Soleimani has sharply escalated tensions with Iran, raising fears of all-out conflict. Washington says it killed Soleimani in self-defense, aiming to disrupt his plans to attack U.S. personnel and interests.

Secretary of State Mike Pompeo scoffed Tuesday at a suggestion that Iranian Gen. Qassem Soleimani was on a diplomatic mission to Baghdad when a U.S. drone strike killed him, insisting that the story is "fundamentally false" and "propaganda." "Anybody here believe that?" Pompeo asked reporters when the claim was brought up during a Tuesday press briefing. "We know that wasn't true." The notion that Soleimani was on some sort of peace mission was first floated by Iraqi Prime Minister Adil Abdul-Mahdi, who reportedly said at a Sunday session he was scheduled to meet with Soleimani and discuss a message from Saudi Arabia regarding possible agreements in the region. Pompeo addressed the claim that Soleimani was supposed to work on a peace agreement with Saudi Arabia, stating that he spoke to Saudi officials about this.

WASHINGTON – The Pentagon on Monday distanced itself from U.S. President Donald Trump's assertions that he would bomb Iranian cultural sites despite international prohibitions on such attacks. Defense Secretary Mark Esper said the U.S. will "follow the laws of armed conflict." When asked if that ruled out targeting cultural sites, Esper said pointedly, "That's the laws of armed conflict." The split between the president and his Pentagon chief came amid heightened tensions with Tehran following a U.S. drone strike that killed Gen. Qassem Soleimani, the head of Iran's elite Quds Force. Trump had twice warned that he would hit Iranian cultural sites if Tehran retaliates against the U.S. Esper's public comments reflected the private concerns of other defense and military officials, who cited legal prohibitions on attacks on civilian, cultural and religious sites, except under certain, threatening circumstances.