cs.CVApr 4, 2026

Determined by User Needs: A Salient Object Detection Rationale Beyond Conventional Visual Stimuli

Authors: Chenglizhao Chen, Shujian Zhang, Luming Li, Wenfeng Song, Shuai Li

Organizations: China University of Petroleum (East China) · Beijing Information Science and Technology University · Beihang University

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Existing salient object detection (SOD) methods adopt a passive visual stimulus-based rationale--objects with the strongest visual stimuli are perceived as the user's primary focus (i.e., salient objects). They ignore the decisive role of users' proactive needs in segmenting salient objects--if a user has a need before seeing an image, the user's salient objects align with their needs, e.g., if a user's need is white apple'', when this user sees an image, the user's primary focus is on the white apple'' or ``the most white apple-like'' objects in the image. Such an oversight not only fails to satisfy users, but also limits the development of downstream tasks. For instance, in salient object ranking tasks, focusing solely on visual stimuli-based salient objects is insufficient for conducting an analysis of fine-grained relationships between users' viewing order (usually determined by user's needs) and scenes, which may result in wrong ranking results. Clearly, it is essential to detect salient objects based on user needs. Thus, we advocate a User Salient Object Detection (UserSOD) task, which focuses on detecting salient objects align with users' proactive needs when user have needs. The main challenge for this new task is the lack of datasets for model training and testing.

Figures & tables

Figure 1: Motivation demonstration of our new task. Current SOTA methods (A) always segment salient object based on passive visual stimulus-based rationales, limiting the performance of downstream tasks ( e.g. , salient object ranking, see details in the Fig. 2 ). Therefore, we introduce a novel task, UserSOD (B), which detects salient object based on user need, as it accounts for the crucial role of user active need, thereby aligning more closely with real users’ primary focus mechanisms.
Figure 2: Application of salient object detection method to salient object ranking (SOR). (A) The existing SOTA SOD methods ignores user active need, leading to constant rank objects’ saliency that fail to satisfy the user needs and deviate from the ground truth (GT, Bottom left). In contrast, our approach (B) can return more fine-grained rank results, better meeting user needs.
Figure 3: The pipeline of our method. Our consists of two components, i.e. , user need digger (UND) and user need-driven salient object detection model (Uersal + ). UND semi-automatically digs user need commands based on existing samples, and Uersal + can leverage fine-grained clues from user need commands to detect salient objects.
Figure 4: Existing Sets v.s. Our UserSOD set. Compared to existing set, Users contains samples which are image, corresponding user need commands, and corresponding g round t ruths (GT). Where, i ⊂\subset (1, +∞+\infty ) and j ⊂\subset (1, +∞+\infty ) denote the number of mask and user need command in single samples, respectively.
Figure 5: The proposed User Need Digger.
Figure 6: The proposed user need-driven model: Usersal (A) and Usersal + (B).
Figure 7: Visualization of similar feature.
Table 1: Quantitative comparisons between our method with other SOTA models on our proposed UserSOD set. The top three precision are highlighted with red, green and blue fonts.
Table 2: Quantitative comparisons between our method with SOTA SOD models on the conventional SOD sets. The top three precision are highlighted with red, green and blue fonts.
Figure 8: Visual comparisons between our SOTA v.s . SOTA SOD and RIS methods. Our method not only achieves sharp contours for conventional SOD but also meets fine-grained user needs.
Table 3: Our method’s effectiveness for downstream SOR task. ∗ and Ave. denote baselines using our methods and average user satisfaction score, respectively.
Table 4: Generalization evaluation on RIS Sets and user study. The top three precision are highlighted with red, green and blue fonts. Where, Ave. and mIOU denotes average user satisfaction score and widely-used metric on RIS sets, respectively.
Table 5: Component evaluation on conventional SOD and UserSOD sets.
Table 6: Generalization evaluation for different baselines on conventional SOD and UserSOD sets. ∗ denote baselines using our methods.
Table 7: Different command encoder evaluation on conventional SOD and UserSOD sets.
Table 8: Tiny siamese network evaluation on conventional SOD and UserSOD sets. ∗ denotes single-layers.
More in this topic

Salient Object Detection

Salient Object Detection finds visually prominent objects in images or video and predicts their pixel-level foreground masks.

See all 37 papers

Explore similar work

CardsList

  1. TDSal: Task-Based Top-Down Saliency Prediction Model

    Jul 10, 2026Can Mizrakli, Tolga K. CapinVisual AttentionSalient Object Detection

    Karlsruhe Institute of Technology, Germany · TED University, Türkiye

  2. Rank-Consistent Set Reasoning for Co-Salient Object Detection

    Sep 12, 2026Yuan Xiang, Matteo Rossi, Yingzhou ChenSalient Object Detection

    University of California, Los Angeles (UCLA) USA · Polytechnic University of Turin Italy

  3. Is It Time for the Renaissance of Salient Object Detection in the Era of MLLMs?

    Jul 31, 2026Wenzhuo Zhao, Xiuzhi Li, Zhongkuan Mao +6VLM EvaluationSalient Object Detection

    College of Computer Science, Sichuan University, China · National Key Laboratory of Fundamental Science on Synthetic Vision, Sichuan University, China · School of Information and Communication Engineering, University of ElectronicScience and Technology of China, China