Image and Vision Computing

By Christos Sagonas, Epameinondas Antonakos, Georgios Tzimiropoulos, Stefanos P. Zafeiriou, Maja Pantic. 2016

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Logic Tensor Networks for Semantic Image Interpretation (2017, neutral): “of IJCAI 2017 The major obstacle that need to be overcome in SII is the so-called semantic gap [Neumann and Möller, 2008], that is the lack of a direct correlation between low-level features of the image and high-level semantic concepts.”
Unsupervised Inductive Graph-Level Representation Learning via Graph-Graph Proximity (2019, background): “This should not be a surprise, because only UGRAPHEMB utilizes the ground-truth GED results collectively determined by BEAM [Neuhaus et al., 2006], HUNGARIAN [Riesen and Bunke, 2009], and VJ [Fankhauser et al., 2011].”
GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians (2023, background): “Additionally, we lack control over areas that are not modeled by FLAME such as hair and other accessories.”
Generalizing Face Forgery Detection with High-frequency Features (2021, methodology): “We use DLIB [41] for face extraction and alignment and resize the aligned faces to 256 × 256.”
One-Shot Generalization in Deep Generative Models (2016, methodology): “The Multi-PIE dataset (Gross et al., 2010) consists of 48 × 48 RGB face images from various viewpoints.”
Style Aggregated Network for Facial Landmark Detection (2018, methodology): “We also use Cumulative Error Distribution (CED) curve to compare the algorithms provided in [45].”
AgeDB: The First Manually Collected, In-the-Wild Age Database (2017, methodology): “Therefore, their N = 68 facial landmarks points are localized using the face detector from [1], trained with images provided from 300-W challenge [44, 41, 43] and subsequently employed to compute a mean reference shape.”
AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time (2022, methodology): “MPII [15] 16 ✓ ✓ 40K CrowdPose [19] 14 ✓ ✓ 80K PoseTrack [68] 15 ✓ ✓ 150K COCO [14] 17 ✓ ✓ 250K OneHand10K [70] 21 ✓ ✓ 10K FreiHand [62] 21 ✓ 130K MHP [71] 21 ✓ 80K WFLW [72] 98 ✓ ✓ 10K AFLW [73] 19 ✓ ✓ 25K COFW [74] 29 ✓ ✓ 1852 300W [61] 68 ✓ ✓ 3837 COCO-WholeBody [18] 133 ✓ ✓ ✓ ✓ 250K Halpe-FullBo”
3FabRec: Fast Few-Shot Face Alignment by Reconstruction (2019, methodology): “300-W This dataset was assembled by [40] from several sources, including LFPW [2], AFW [26], HELEN [61], XM2VTS [32], and own data and annotated semiautomatically with 68 facial landmarks.”
Mnemonic Descent Method: A Recurrent Process Applied for End-to-End Face Alignment (2016, methodology): “CLMs have not achieved state-of-the-art results in recent challenges such as the 300-W competition [38, 39], even when trained in a cascaded regression [4].”
Unsupervised Learning of Object Landmarks by Factorized Spatial Embeddings (2017, methodology): “We also consider the more challenging 300-W dataset [54] containing 68 landmarks, obtained by merging and re-annotating other benchmarks.”
3D Face Morphable Models "In-the-Wild" (2017, methodology): “300W datasets [31, 32] that qualitatively highlight the effectiveness of the proposed technique.”