3D scene understanding from point clouds plays a vital role for various robotic applications. Unfortunately, current state-of-the-art methods use separate neural networks for different tasks like object detection or room layout estimation. Such a sch
eme has two limitations: 1) Storing and running several networks for different tasks are expensive for typical robotic platforms. 2) The intrinsic structure of separate outputs are ignored and potentially violated. To this end, we propose the first transformer architecture that predicts 3D objects and layouts simultaneously, using point cloud inputs. Unlike existing methods that either estimate layout keypoints or edges, we directly parameterize room layout as a set of quads. As such, the proposed architecture is termed as P(oint)Q(uad)-Transformer. Along with the novel quad representation, we propose a tailored physical constraint loss function that discourages object-layout interference. The quantitative and qualitative evaluations on the public benchmark ScanNet show that the proposed PQ-Transformer succeeds to jointly parse 3D objects and layouts, running at a quasi-real-time (8.91 FPS) rate without efficiency-oriented optimization. Moreover, the new physical constraint loss can improve strong baselines, and the F1-score of the room layout is significantly promoted from 37.9% to 57.9%.
Let $mathscr C$ be a Krull-Schmidt $(n+2)$-angulated category and $mathscr A$ be an $n$-extension closed subcategory of $mathscr C$. Then $mathscr A$ has the structure of an $n$-exangulated category in the sense of Herschend-Liu-Nakaoka. This constru
ction gives $n$-exangulated categories which are not $n$-exact categories in the sense of Jasso nor $(n+2)$-angulated categories in the sense of Geiss-Keller-Oppermann in general. As an application, our result can lead to a recent main result of Klapproth.
A notion of balanced pairs in an extriangulated category with a negative first extension is defined in this article. We prove that there exists a bijective correspondence between balanced pairs and proper classes $xi$ with enough $xi$-projectives and
enough $xi$-injectives. It can be regarded as a simultaneous generalization of Fu-Hu-Zhang-Zhu and Wang-Li-Huang. Besides, we show that if $(mathcal A ,mathcal B,mathcal C)$ is a recollement of extriangulated categories, then balanced pairs in $mathcal B$ can induce balanced pairs in $mathcal A$ and $mathcal C$ under natural assumptions. As a application, this result gengralizes a result by Fu-Hu-Yao in abelian categories. Moreover, it highlights a new phenomena when it applied to triangulated categories.
Let $n$ be an integer greater or equal than $3$. We give a simultaneous generalization of $(n-2)$-exact categories and $n$-angulated categories, and we call it one-sided $n$-suspended categories. One-sided $n$-angulated categories are also examples o
f one-sided $n$-suspended categories. We provide a general framework for passing from one-sided $n$-suspended categories to one-sided $n$-angulated categories. Besides, we give a method to construct $n$-angulated quotient categories from Frobenius $n$-prile categories. These results generalize their works by Jasso for $n$-exact categories, Lin for $(n+2)$-angulated categories and Li for one-sided suspended categories.
Recently, Wang, Wei and Zhang define the recollement of extriangulated categories, which is a generalization of both recollement of abelian categories and recollement of triangulated categories. For a recollement $(mathcal A ,mathcal B,mathcal C)$ of
extriangulated categories, we show that $n$-tilting (resp. $n$-cotilting) subcategories in $mathcal A$ and $mathcal C$ can be glued to get $n$-tilting (resp. $n$-cotilting) subcategories in $mathcal B$ under certain conditions.
The notion of right semi-equivalence in a right $(n+2)$-angulated category is defined in this article. Let $mathscr C$ be an $n$-exangulated category and $mathscr X$ is a strongly covariantly finite subcategory of $mathscr C$. We prove that the stand
ard right $(n+2)$-angulated category $mathscr C/mathscr X$ is right semi-equivalence under a natural assumption. As an application, we show that a right $(n+2)$-angulated category has an $n$-exangulated structure if and only if the suspension functor is right semi-equivalence. Besides, we also prove that an $n$-exangulated category $mathscr C$ has the structure of a right $(n+2)$-angulated category with right semi-equivalence if and only if for any object $Xinmathscr C$, the morphism $Xto 0$ is a trivial inflation.
We introduce pre-silting and silting subcategories in extriangulated categories and generalize the silting theory in triangulated categories. We prove that the silting reduction $mathcal B/({rm thick}mathcal W)$ of an extriangulated category $mathcal
B$ with respect to a pre-silting subcategory $mathcal W$ can be realized as a certain subfactor category of $mathcal B$. This generalizes the result by Iyama-Yang. In particular, for a Gorenstein algebra, we get the relative version of the description of the singularity category due to Happel and Chen-Zhang by this reduction.
We introduce a new concept of s-recollements of extriangulated categories, which generalizes recollements of abelian categories, recollements of triangulated categories, as well as recollements of extriangulated categories. Moreover, some basic prope
rties of s-recollements are presented. We also discuss the behavior of the localization theory on the adjoint pair of exact functors. Finally, we provide a method to obtain a recollement of triangulated categories from s-recollements of extriangulated categories via the localization theory.
Tissue deformation in ultrasound (US) imaging leads to geometrical errors when measuring tissues due to the pressure exerted by probes. Such deformation has an even larger effect on 3D US volumes as the correct compounding is limited by the inconsist
ent location and geometry. This work proposes a patient-specified stiffness-based method to correct the tissue deformations in robotic 3D US acquisitions. To obtain the patient-specified model, robotic palpation is performed at sampling positions on the tissue. The contact force, US images and the probe poses of the palpation procedure are recorded. The contact force and the probe poses are used to estimate the nonlinear tissue stiffness. The images are fed to an optical flow algorithm to compute the pixel displacement. Then the pixel-wise tissue deformation under different forces is characterized by a coupled quadratic regression. To correct the deformation at unseen positions on the trajectory for building 3D volumes, an interpolation is performed based on the stiffness values computed at the sampling positions. With the stiffness and recorded force, the tissue displacement could be corrected. The method was validated on two blood vessel phantoms with different stiffness. The results demonstrate that the method can effectively correct the force-induced deformation and finally generate 3D tissue geometries
In the Gaia era, the membership analysis and parameter determination of open clusters (OCs) are more accurate. We performed a census of OCs classical Cepheids based on Gaia Early Data Release 3 (EDR3) and obtained a sample of 33 OC Cepheids fulfillin
g the constraints of the spatial position, proper motion, parallax and evolution state. 13 of 33 OC Cepheids are newly discovered. Among them, CM Sct is the first first-crossing Cepheids with direct evidence of evolution. DP Vel is likely a fourth- or fifth-crossing Cepheids. Based on independent distances from OCs, W_1-band period-luminosity relation of Cepheids is determined with a 3.5% accuracy: <MW1> = -(3.274 +- 0.090) log P - (-2.567 +- 0.080). The Gaia-band period-Wesenheit relation agrees well with Ripepi et al. (2019). A direct period-age relation for fundamental Cepheids are also determined based on OCs age, that is log t = -(0.638 +- 0.063) log P + (8.569 +- 0.057).
