Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386075900> ?p ?o ?g. }
- W4386075900 abstract "Recent 3D face reconstruction methods have made significant advances in geometry prediction, yet further cosmetic improvements are limited by lagged albedo because inferring albedo from appearance is an ill-posed problem. Although some existing methods consider prior knowledge from illumination to improve albedo estimation, they still produce a light-skin bias due to racially biased albedo models and limited light constraints. In this paper, we reconsider the relationship between albedo and face attributes and propose a ID2Albedo to directly estimate albedo without constraining illumination. Our key insight is that intrinsic semantic attributes such as race, skin color, and age can be used to constrain the albedo map. We first introduce visual-textual cues and design a semantic loss to supervise facial albedo estimation. Specifically, we pre-define text labels such as race, skin color, age, and wrinkles. Then, we employ the text-image model (CLIP) to compute the similarity between the text and the input image, and assign a pseudo-label to each facial image. We constrain generated albedos in the training phase to have the same attributes as the inputs. In addition, we train a high-quality, unbiased facial albedo generator and utilize the semantic loss to learn the mapping from illumination-robust identity features to the albedo latent codes. Finally, our ID2Albedo is trained in a self-supervised way and outperforms state-of-the-art albedo estimation methods in terms of accuracy and fidelity. It is worth mentioning that our approach has excellent generalizability and fairness, especially on in-the- wild data." @default.
- W4386075900 created "2023-08-23" @default.
- W4386075900 creator A5019708391 @default.
- W4386075900 creator A5040299928 @default.
- W4386075900 creator A5048907728 @default.
- W4386075900 creator A5051172458 @default.
- W4386075900 creator A5073110676 @default.
- W4386075900 date "2023-06-01" @default.
- W4386075900 modified "2023-10-16" @default.
- W4386075900 title "Improving Fairness in Facial Albedo Estimation via Visual-Textual Cues" @default.
- W4386075900 cites W1566555074 @default.
- W4386075900 cites W2003706019 @default.
- W4386075900 cites W2099478163 @default.
- W4386075900 cites W2107037917 @default.
- W4386075900 cites W2108492831 @default.
- W4386075900 cites W2120420721 @default.
- W4386075900 cites W2194775991 @default.
- W4386075900 cites W2237250383 @default.
- W4386075900 cites W2404498690 @default.
- W4386075900 cites W2486034530 @default.
- W4386075900 cites W2547473593 @default.
- W4386075900 cites W2565639579 @default.
- W4386075900 cites W2772024431 @default.
- W4386075900 cites W2786166435 @default.
- W4386075900 cites W2798291180 @default.
- W4386075900 cites W2962770929 @default.
- W4386075900 cites W2962785568 @default.
- W4386075900 cites W2964449965 @default.
- W4386075900 cites W2969985801 @default.
- W4386075900 cites W2982232682 @default.
- W4386075900 cites W2990431681 @default.
- W4386075900 cites W3034192160 @default.
- W4386075900 cites W3034552680 @default.
- W4386075900 cites W3034771054 @default.
- W4386075900 cites W3035413889 @default.
- W4386075900 cites W3035574324 @default.
- W4386075900 cites W3101531717 @default.
- W4386075900 cites W3159093935 @default.
- W4386075900 cites W3163977797 @default.
- W4386075900 cites W3166898278 @default.
- W4386075900 cites W3167584510 @default.
- W4386075900 cites W3180355996 @default.
- W4386075900 cites W3180770160 @default.
- W4386075900 cites W3180794345 @default.
- W4386075900 cites W3181238075 @default.
- W4386075900 cites W3201927751 @default.
- W4386075900 cites W3203983723 @default.
- W4386075900 cites W4214926101 @default.
- W4386075900 cites W4250928742 @default.
- W4386075900 cites W4286611278 @default.
- W4386075900 cites W4313041309 @default.
- W4386075900 doi "https://doi.org/10.1109/cvpr52729.2023.00438" @default.
- W4386075900 hasPublicationYear "2023" @default.
- W4386075900 type Work @default.
- W4386075900 citedByCount "0" @default.
- W4386075900 crossrefType "proceedings-article" @default.
- W4386075900 hasAuthorship W4386075900A5019708391 @default.
- W4386075900 hasAuthorship W4386075900A5040299928 @default.
- W4386075900 hasAuthorship W4386075900A5048907728 @default.
- W4386075900 hasAuthorship W4386075900A5051172458 @default.
- W4386075900 hasAuthorship W4386075900A5073110676 @default.
- W4386075900 hasConcept C105795698 @default.
- W4386075900 hasConcept C115961682 @default.
- W4386075900 hasConcept C142362112 @default.
- W4386075900 hasConcept C144024400 @default.
- W4386075900 hasConcept C154945302 @default.
- W4386075900 hasConcept C195886398 @default.
- W4386075900 hasConcept C27158222 @default.
- W4386075900 hasConcept C2779304628 @default.
- W4386075900 hasConcept C31972630 @default.
- W4386075900 hasConcept C33923547 @default.
- W4386075900 hasConcept C36289849 @default.
- W4386075900 hasConcept C41008148 @default.
- W4386075900 hasConcept C52119013 @default.
- W4386075900 hasConcept C554144382 @default.
- W4386075900 hasConceptScore W4386075900C105795698 @default.
- W4386075900 hasConceptScore W4386075900C115961682 @default.
- W4386075900 hasConceptScore W4386075900C142362112 @default.
- W4386075900 hasConceptScore W4386075900C144024400 @default.
- W4386075900 hasConceptScore W4386075900C154945302 @default.
- W4386075900 hasConceptScore W4386075900C195886398 @default.
- W4386075900 hasConceptScore W4386075900C27158222 @default.
- W4386075900 hasConceptScore W4386075900C2779304628 @default.
- W4386075900 hasConceptScore W4386075900C31972630 @default.
- W4386075900 hasConceptScore W4386075900C33923547 @default.
- W4386075900 hasConceptScore W4386075900C36289849 @default.
- W4386075900 hasConceptScore W4386075900C41008148 @default.
- W4386075900 hasConceptScore W4386075900C52119013 @default.
- W4386075900 hasConceptScore W4386075900C554144382 @default.
- W4386075900 hasLocation W43860759001 @default.
- W4386075900 hasOpenAccess W4386075900 @default.
- W4386075900 hasPrimaryLocation W43860759001 @default.
- W4386075900 hasRelatedWork W1899364738 @default.
- W4386075900 hasRelatedWork W2018638282 @default.
- W4386075900 hasRelatedWork W2103413230 @default.
- W4386075900 hasRelatedWork W2116300362 @default.
- W4386075900 hasRelatedWork W2138569648 @default.
- W4386075900 hasRelatedWork W2143020626 @default.
- W4386075900 hasRelatedWork W2663901905 @default.
- W4386075900 hasRelatedWork W2908959303 @default.